Keywords: Automated Vehicles
Abstract: Intelligent Connected Vehicles Based on Advanced Communication Technologies

https://drive.google.com/file/d/17W8adQ2gRD_8hrJdVGiI4oyFTM NSb2D1/view?usp=sharing

Keywords: Cooperative Vehicles, Automated Vehicles, Integration of Infrastructure and Intelligent Vehicles
Abstract: This paper proposed a right-of-way allocation method for Automated Intelligent Vehicles (AIVs) in automated container terminals. The results demonstrate a notable improvement in travel efficiency, up to 60%, along with responsiveness measured in seconds.

Keywords: Cooperative Vehicles, Automated Vehicles, Vehicle Control and Motion Planning
Abstract: This paper presents a novel consensus control algorithm for a vehicle platoon, taking into account human physiological-psychological comfort. To this end, this paper incorporates state constraints into the optimal velocity-based platoon controller, which can effectively improve the comfort and consensus. Moreover, this paper employs the barrier Lyapunov function (BLF) to prove the stability of the proposed controller, providing more strict stability guarantees. According to theoretical analysis, numerical experiments are conducted to demonstrate the performance of the proposed control algorithm and the benchmark algorithm. The comparative results show that the proposed control algorithm can produce better-uniformed states and bounded state errors.

Keywords: Human Factors for Intelligent Vehicles, Advanced Driver Assistance Systems (ADAS), Simulation and Real-World Testing Methodologies
Abstract: Analyzing discretionary lane-changing (DLC) behavior can provide support for intelligent connected vehicle driving behavior modeling and microscopic traffic simulation. The discrete choice model is the basic model in DLC modeling. However, the discrete choice model cannot fully consider risk preferences during the DLC decision making process. This study harnesses cumulative prospect theory to consider risk preference, constructing a feature vector that includes lane-changing benefits, lane-changing risks, and driving style of drivers. Based on this, a DLC decision making framework based on the discrete choice model is proposed. The framework considers risk preference in the DLC decision making process. It not only retains the interpretability of the discrete choice model but also enhances the predictive performance of DLC. The DLC decision-making framework proposed in this study is validated through the Next Generation Simulation (NGSIM) dataset. The results show that the DLC decision making framework proposed in this study can achieve better performance than the discrete choice model, with an accuracy reaching 77.25% in the test set.

Keywords: Vehicle Control and Motion Planning, Cooperative Vehicles, Automated Vehicles
Abstract: This paper introduces an online physical enhanced residual learning (PERL) framework for Connected Autonomous Vehicles (CAVs) platoon, aimed at addressing the challenges posed by the dynamic and unpredictable nature of traffic environments. The proposed framework synergistically combines a physical model, represented by Model Predictive Control (MPC), with data-driven online Q-learning. The MPC controller, enhanced for centralized CAV platoons, employs vehicle velocity as a control input and focuses on multi-objective cooperative optimization. The learning-based residual controller enriches the MPC with prior knowledge and corrects residuals caused by traffic disturbances. The PERL framework not only retains the interpretability and transparency of physics-based models but also significantly improves computational efficiency and control accuracy in real-world scenarios. The experimental results present that the online Q-learning PERL controller, in comparison to the MPC controller and PERL controller with a neural network, exhibits significantly reduced position and velocity errors. Specifically, the PERL's cumulative absolute position and velocity errors are, on average, 86.73% and 55.28% lower than the MPC's, and 12.82% and 18.83% lower than the neural network-based PERL's, in four tests with different reference trajectories and errors. The results demonstrate our advanced framework's superior accuracy and quick convergence capabilities, proving its effectiveness in maintaining platoon stability under diverse conditions.

Keywords: Automated Vehicles
Abstract: Workshop on Remote Operation of Intelligent Connected and Automated Road Vehicles

Website: https://www.vti.se/rvt/

This workshop intends to serve as a common research arena to initiate multi-disciplinary discussions on different components around remote operation of intelligent connected and automated vehicles on the road. This is the fourth workshop following the first three workshops on the same topic at IV2021, IV2022, and IV2023 (see below for info on previous workshops).

Road vehicles of interest include (but are not limited to) connected and automated vehicles that are passenger cars, trucks, and shuttles or buses. In this context, we assume there is a remote operator who operates the vehicle from a distance via wireless communication network. The remote operation is done at a remote operation station, where necessary interfaces for remotely operating the vehicle(s) are provided.

Keywords: Teleoperation of Intelligent Vehicles, Human Factors for Intelligent Vehicles
Abstract: Remote driving is a promising strategy for helping Autonomous Vehicles (AVs) navigate many environments where edge cases may otherwise limit their abilities. For some companies, remote driving is an alternative to AVs altogether. Much remote driving research has taken place in simulated or controlled environments with novice operators, leaving the needs of operators with real-world experience under-explored. This research aims to understand if experienced operators are satisfied with current production remote driving systems, if they adapt to the difference in control, and how their job satisfaction compares to in-vehicle safety driving. This paper briefly overviews recent remote driving research and presents results from a questionnaire and a semi-structured interview with experienced teleoperators. The findings indicate that operators do adjust to the new domain, but latency and network reliability remain a challenge. Likewise, standardised training practices for operators are found to be lacking.

Keywords: Teleoperation of Intelligent Vehicles, Human Factors for Intelligent Vehicles
Abstract: Despite the rapid technological progress, autonomous vehicles still face a wide range of complex driving situations that require human intervention. Teleoperation technology offers a versatile and effective way to address these challenges. The following work puts existing ideas into a modern context and introduces a novel technical implementation of the trajectory guidance teleoperation concept. The presented system was developed within a high-fidelity simulation environment and experimentally validated, demonstrating a realistic ride-hailing mission with prototype autonomous vehicles and onboard passengers. The results indicate that the proposed concept can be a viable alternative to the existing remote driving options, offering a promising way to enhance teleoperation technology and improve overall operation safety.

Keywords: Teleoperation of Intelligent Vehicles, Human Factors for Intelligent Vehicles, Automated Vehicles
Abstract: The remote operation of automated vehicles stands as a pivotal technological solution for bridging the current driving automation technology toward the realization of safe and efficient mobility services. In the context of passenger vehicles, such as driverless buses, a remote operator will need to switch attention among different vehicles and tasks, which can include both driving-related and passenger-service related tasks. The potential cost associated with such switching may impact the operator’s performance in providing effective remote support. This study aimed to investigate the effect of task-switching costs in the remote operation of automated vehicle fleet. In an experiment involving 60 participants spanning three age groups, participants were instructed to perform remote operation tasks under three task-switching conditions. The results revealed that although a constant switch between two tasks did not affect the remote operator’s performance, there was a significant slowdown when participants randomly switched between three different tasks. Furthermore, older participants reacted significantly more slowly than younger and middle-aged participants in performing the remote operation tasks. These findings emphasized the necessity of considering the time required for remote operators to shift their attention when determining an optimal human-to-vehicle ratio. Additionally, although older participants reacted more slowly in performing the operation tasks, their response accuracies were comparable to younger and middle-aged participants, indicating that they remain suited to serve as remote operators as long as their slightly slower pace is accommodated.

Keywords: Teleoperation of Intelligent Vehicles, Integration of Infrastructure and Intelligent Vehicles, Vehicle-To-Everything (V2X) and Cellular V2X (C-V2X) Communications
Abstract: As autonomous vehicles become increasingly prevalent, robust remote operation systems are imperative to ensure safety and reliability in unpredictable scenarios. Current remote operation systems in research often lack scalability and adaptability, hindering their integration into diverse autonomous driving platforms. This paper addresses these challenges by introducing a scalable remote operation system that leverages cooperative perception and an open-source communication module. Field tests conducted with an SAE Level 3 autonomous shuttle have validated the effectiveness of our system in real-world scenarios. The code for a key component of this system, the communication module, is available online: https://github.com/fzi-forschungszentrum-informatik/ros_communication_devcontainer.

Keywords: Teleoperation of Intelligent Vehicles, Vehicle-To-Everything (V2X) and Cellular V2X (C-V2X) Communications, Functional Safety in Intelligent Vehicles
Abstract: Remote driving plays a vital role in coordinating automated vehicles in challenging situations. Data transmission latency, however, can cause several problems in remote driving. Firstly, it can degrade the performance of remote-controlled vehicles, evident in issues like lane-following deviation and vehicle stability. Additionally, the remote control tower's driving feedback is affected by delayed vehicle signals, leading to delayed driving experience. To address this, a model-free-based predictor is employed to compensate for the delay in remote driving. This approach does not require any dynamic model of the system and only needs tuning of two parameters to reduce communication delay. This study enhances the previous work by mitigating the amplitude of overshoot around peak points. It leverages the principle of the second-order derivative to predict the signal's peak time and uses it to address the predictor's overshoot issue. The effectiveness of the proposed method is validated using real car data from multiple participants in two scenarios, including Slalom and lane-following. Simulation results indicate that the proposed method can reduce prediction error by nearly 25% compared to previous works. Moreover, the solutions in this study are capable of managing not only delays in remote driving vehicles but also in traditional mechanical systems, such as CAN bus delays in conventional cars.

Keywords: Automated Vehicles, Teleoperation of Intelligent Vehicles, Future Mobility and Smart City
Abstract: Remote operators (ROs) of automated vehicles will be unable to provide remote assistance until they gain necessary situation awareness (SA). A critical question for the industry concerns the optimal format to deliver information to an RO. In this study, we consider remote assistance of automated vehicles using a choice decision task to test the effect of two formats of presentation of 360° driving videos: 1) in a head mounted display (HMD-360) where field of view (FOV) changes were based on head movement and 2) screen-based (SB) mouse-controlled FOV. Participants viewed 60 videos depicting scenarios categorised into seven groups, each representing different types of edge cases commonly encountered in real-world situations by ROs and distributed across various decision choices (left, right, forward, or reverse). Decision time and accuracy of decision were recorded. Analysis revealed significantly quicker decisions and higher accuracy in the HMD-360 condition. We recommend exploring HMD-360 presentation to improve operator SA in remote assistance of AVs, while cautioning against prolonged HMD use to mitigate potential discomfort.

Keywords: Automated Vehicles
Abstract: Workshop on Modular Autonomous Vehicles (MAVs)

https://ieeeiv2024.weebly.com/

Modular Autonomous Vehicles (MAVs) can revolutionize the way people and goods are transported in various urban and industrial settings. This workshop aims to bring academic and industrial scientists and researchers to share their knowledge and experiences in MAVs research, development and applications. Original research and review articles related to the planning, design, operation, control, management, maintenance, and technological aspects of MAVs will be considered. All aspects of mathematical modelling, computer simulation, statistical analysis, economic analysis, and empirical studies are of interest. We particularly welcome submissions with a focus on the implementation and field testing of new methods and technologies for MAVs.​

Keywords: Future Mobility and Smart City, Wireless Power Transfer Systems for Mobility
Abstract: The mobile vehicle-to-vehicle charging vehicles (MCV) for electric modular buses (EMB) have greatly advanced in recent years, where the MCV is an electric vehicle with an extra battery that can charge another electric vehicle on the move. In this study, we propose a mobile vehicle routing problem for electric modular buses, in which both MCVs and charging stations are used to charge electric modular buses. This study proposes a joint optimization model for EMBs and MCVs, where the CPLEX solver was used to solve the constructed case to verify the feasibility and validity of the proposed model, and the results show a 10.5% overall cost reduction compared to the traditional electric modular bus system. Meanwhile, sensitivity tests are designed to reveal the effects of the ratio of charge rate ratio and the number of serviced trips on the performance of the electric modular bus system.

Keywords: Automated Vehicles
Abstract: Workshop: SAFE-DRIVE: Data-Driven Simulations and Multi-Agent Interactions for Autonomous Vehicle Safety

Workshop Website : https://safeav.github.io/

Keywords: Human Factors for Intelligent Vehicles
Abstract: Since future mobility will change into mixed traffic, the communication approach of automated vehicles should consider multiple human receivers. This virtual reality study investigates the external communication of an automated vehicle in interactions with multiple pedestrians. Involving 42 participants, the study explores the impact of a simulated pedestrian group at the roadside, a seldom-explored aspect in current research on human-vehicle interactions. Results reveal a significant influence of the additional pedestrian group on participants' behaviors and perceptions, emphasizing the need for nuanced communication strategies in automated vehicle design. This paper contributes to understanding social factors in automated vehicle interactions, providing valuable insights for future research and the development of human-centric automated vehicle communication systems.

Keywords: Pedestrian Protection, Human Factors for Intelligent Vehicles, Functional Safety in Intelligent Vehicles
Abstract: Autonomous vehicles face significant challenges in understanding pedestrian behavior, particularly in urban environments. In such settings, the system must recognize pedestrian intentions and anticipate their actions to achieve safe and intelligent driving. This paper focuses on predicting pedestrian crossings, enabling oncoming vehicles to react to pedestrians in a traffic scene in a timely manner. We investigate the effectiveness of various input features for pedestrian crossing prediction, including human poses, bounding boxes, and ego vehicle speed features. We propose a novel lightweight architecture based on LSTM and attention to accurately identify crossing pedestrians. Our methods are evaluated on two widely used public datasets for pedestrian behavior, PIE and JAAD datasets, and our algorithm achieves a state-of-the-art performance in both datasets by reaching a prediction accuracy of 91% and an F1-score of 84% on the PIE dataset and an accuracy of 67% and an F1-score of 77% on the JAAD Behavior. We make our code available at https://github.com/afnan29alofi/Pedestrian-Intent-Predictio n.git

Keywords: Simulation and Real-World Testing Methodologies, Human Factors for Intelligent Vehicles, Pedestrian Protection
Abstract: Virtual Reality (VR) is commonly utilized to examine driver interactions with vulnerable road users (VRUs) in an effective and secure manner. Recent studies, however, have highlighted issues in VR simulations, particularly concerning the authenticity of state-of-the-art pedestrian agent behaviors. These inaccuracies can compromise the perceived realism of the situation, potentially leading to unrepresentative driver reactions. This paper aims to showcase enhancements in pedestrian agent models and evaluate their subsequent advantages. To this end, real pedestrian movements, captured via motion-capture technology, were compared with outputs from a contemporary pedestrian agent model within a VR driving simulator experiment. The findings underpin the advantages of using motion-captured pedestrians to enhance social presence. Additionally, participant feedback emphasized that certain elements, such as head movements, explicit gestures, and subtle cues like hesitation before entering the road, were crucial in distinguishing realistic from unrealistic agents. These insights contribute significantly to refining the focus for systematic advancements in (pedestrian) agent models in VR environments. Such improvements are pivotal in augmenting the users' sense of presence and the behavioral accuracy of the simulations.

Keywords: Simulation and Real-World Testing Methodologies, Vehicle Control and Motion Planning, Vehicle-To-Everything (V2X) and Cellular V2X (C-V2X) Communications
Abstract: Traffic simulation is important for transportation researchers, analysts, and policymakers. It can be used to test vehicle/traffic control algorithms, gain insights into traffic dynamics, and develop traffic management strategies that can improve the efficiency and safety of transportation systems. Unfortunately, many existing simulation platforms have limitations to cater to diverse simulation scales. This study presents a comprehensive multiscale vehicle-traffic-demand (VTD) simulation platform tailored for connected and automated transportation systems. This platform integrates Unity 3D, Simulation of Urban Mobility (SUMO), and Multiagent Transport Simulation (MATSim) to facilitate an in-depth analysis of both micro and macro-level traffic behaviors, with a usage example on validating control algorithms for connected and automated vehicles (CAVs). A critical aspect of our work involves the meticulous setup and calibration of traffic networks in Greater and Downtown Seattle, ensuring effective integration and communication between the various simulation tools. This advanced platform not only serves as a robust tool for testing and refining vehicle/traffic control algorithms but also opens new avenues for research into traffic dynamics learning and the development of sophisticated traffic control solutions

Keywords: Automated Vehicles
Abstract: 5th Workshop on Data Driven Intelligent Vehicle Applications (DDIVA)

https://www.ce.cit.tum.de/air/research/ddiva/ddiva24/

This workshop aims to address the challenges in autonomous driving by focusing on Data and Application domains. Well-labeled data is crucial to improve accuracy in deep learning applications. In this workshop, we mainly focus on data and deep learning, since data enables through applications to infer more information about the environment for autonomous driving.

Recent advancements in processing units have improved our ability to construct a variety of architectures for understanding the surroundings of vehicles. Deep learning methods have been developed for geometric and semantic understanding of environments in driving scenarios aiming to increase the success of full autonomy with the cost of large amounts of data. Recently proposed methods challenge this dependency by pre-processing the data, enhancing, collecting, and labeling it intelligently. In addition, the dependency on data can be relieved by generating synthetic data, which alleviates this need with cost-free annotations. This workshop aims to form a platform for exchanging ideas and linking the scientific community active in the intelligent vehicles domain. This workshop will provide an opportunity to discuss applications and their data-dependent demands for the spatiotemporal understanding of the surroundings while addressing how the data can be exploited to improve results instead of changing proposed architectures.

Keywords: Automated Vehicles, Perception Including Object Event Detection and Response (OEDR)
Abstract: Today, deep learning detectors for autonomous driving are delivering impressive results on public datasets and in real-world applications. However, these detectors require large amounts of data, especially labeled data, to achieve the performance needed to ensure safe driving. The process of collecting and tagging data is expensive and cumbersome. Therefore, the recent focus of the industry has been on how to achieve similar performance while limiting the amount of labeled data required to train such models. Within the cross-modal active learning paradigm, we propose and analyze new strategies to exploit the inconsistencies between camera and LiDAR detectors to improve sampling efficiency and label only the samples that bring the most improvement to model training. For this, we leverage the 2D projection of the bounding boxes to equalize the output quality of camera and LiDAR detections. Finally, we achieve up to 0.6% AP improvement for camera and 2% improvement for LiDAR over random sampling on the KITTI dataset using a sampling strategy based on the number of detected boxes.

Keywords: Sensor Signal Processing, Human Factors for Intelligent Vehicles, Advanced Driver Assistance Systems (ADAS)
Abstract: Advanced Driver Assistance Systems (ADAS) in intelligent vehicles rely on accurate driver perception within the vehicle cabin, often leveraging a combination of sensing modalities. However, these modalities operate at varying rates, posing challenges for real-time, comprehensive driver state monitoring. This paper addresses the issue of missing data due to sensor frame rate mismatches, introducing a generative model approach to create synthetic yet realistic thermal imagery. We propose using conditional generative adversarial networks (cGANs), specifically comparing the pix2pix and CycleGAN architectures. Experimental results demonstrate that pix2pix outperforms CycleGAN, and utilizing multi-view input styles, especially stacked views, enhances the accuracy of thermal image generation. Moreover, the study evaluates the model's generalizability across different subjects, revealing the importance of individualized training for optimal performance. The findings suggest the potential of generative models in addressing missing frames, advancing driver state monitoring for intelligent vehicles, and underscoring the need for continued research in model generalization and customization.

Keywords: Automated Vehicles
Abstract: Trajectory planning in autonomous driving is highly dependent on predicting the emergent behavior of other road users. Learning-based methods are currently showing impressive results in simulation-based challenges, with transformer-based architectures technologically leading the way. Ultimately, however, predictions are needed in the real world. In addition to the shifts from simulation to the real world, many vehicle- and country-specific shifts, i.e. differences in sensor systems, fusion and perception algorithms as well as traffic rules and laws, are on the agenda. Since models that can cover all system setups and design domains at once are not yet foreseeable, model adaptation plays a central role. Therefore, a simulation-based study on transfer learning techniques is conducted on basis of a transformer-based model. Furthermore, the study aims to provide insights into possible trade-offs between computational time and performance to support effective transfers into the real world.

Keywords: Automated Vehicles
Abstract: Emerging Role of Smart Infrastructure and UAVs for Transportation Safety and Autonomy

https://asarkar1.github.io/UAV-W/

Surface transportation is changing rapidly. The rapid emergence of recent technologies and their intersection with transportation research has opened new avenues for studies. Smart traffic infrastructures and unmanned aerial vehicles (UAVs) increasingly find their way into research to overcome critical challenges as part of the monitoring, surveillance, and reference measurement system. Intelligent and connected infrastructures are being rolled out for this purpose on proving grounds and test areas in the public domain. In parallel, UAVs equipped with advanced sensing techniques can aid to the performance of transportation infrastructure and safety. They can capture key behaviors of multimodal traffic, infrastructure health, and traffic incident with capacity to communicate with the infrastructure system. This workshop, therefore, aims to bring together experts from relevant fields across the globe to share and discuss the current state of play.

Keywords: Integration of HD map and Onboard Sensors, Automated Vehicles, Functional Safety in Intelligent Vehicles
Abstract: This paper investigates the impact of a corrupted map on the quality of GNSS-free localization. Using a High Definition (HD) map, we gradually remove and shift features in space. From originally 200 features per kilometer, we randomly discard 90% of our map and introduce perturbations of up to +/-0.9m in each direction to our HD map features. Our evaluation is based on a GNSS-free LiDAR-SLAM, which utilizes panoptic segmentation to observe geometric primitives. In the back-end, a graph optimization is performed to estimate the vehicle’s and the landmarks’ position. Further, we also assess the impact of conducting pure localization instead of Simultaneous Localization and Mapping (SLAM). The effects are evaluated using Mean Absolute Error for accuracy evaluation. For stability assessment, we calculate two percentiles: one for deviations of 0.3m or less, and the second for deviations of 2m or less. We conduct real-life experiments with our testing vehicle which is equipped with a reference system based on RTK-GNSS. We use a commercial, standardized HD map of our suburban campus. Our findings aim to provide insights into the trade-offs between mapping effort, map maintenance, and accurate positioning. We demonstrate that mapping the environment enhances the stability of localization.

Keywords: Vehicle-To-Everything (V2X) and Cellular V2X (C-V2X) Communications, Pedestrian Protection, Cooperative Vehicles
Abstract: Recent reports from the World Health Organization highlight that Vulnerable Road Users (VRUs) have been involved in over half of the road fatalities in recent years, with occlusion risk — a scenario where VRUs are hidden from drivers’ view by obstacles like parked vehicles — being a critical contributing factor. To address this, we present a novel algorithm that quantifies occlusion risk based on the dynamics of both vehicles and VRUs. This algorithm has undergone testing and evaluation using a real-world dataset from German intersections. Additionally, we introduce the concept of Maximum Tracking Loss (MTL), which measures the longest consecutive duration a VRU remains untracked by any vehicle in a given scenario. Our study extends to examining the role of the Collective Perception Service (CPS) in VRU safety. CPS enhances safety by enabling vehicles to share sensor information, thereby potentially reducing occlusion risks. Our analysis reveals that a 25% market penetration of CPS-equipped vehicles can substantially diminish occlusion risks and significantly curtail MTL. These findings demonstrate how various scenarios pose different levels of risk to VRUs and how the deployment of Collective Perception can markedly improve their safety. Furthermore, they underline the efficacy of our proposed metrics to capture occlusion risk as a safety factor.

Keywords: Automated Vehicles
Abstract: 1 st Workshop on Small-scale Testbeds for Connected and Automated Vehicles and Robot Swarms

https://cpm-remote.embedded.rwth-aachen.de/iv24-workshop

The design and validation of algorithms for Connected and Automated Vehicles (CAVs) or robot swarms often necessitate computer simulations and/or full-scale experiments. Small-scale testbeds aim to combine the merits of both to enable rapid prototyping while approximating real-world conditions more closely than mere simulations. However, despite their potential, they often face challenges in accessibility, reproducibility and the standardization of best practices. For instance, many testbeds are not easily accessible to researchers, especially those from underrepresented communities. Additionally, there needs to be a platform for effective sharing of hardware designs, software, and data among different testbeds. This workshop strives to address these gaps by establishing a collaborative network among existing testbeds devoted to CAVs or robot swarms. Specifically, the objectives are:

Enhancing Accessibility and Diversity: Initiate discussions and concrete steps to improve testbed accessibility, particularly for underrepresented communities. Sharing Best Practices: Create a platform for discussing best practices in sustaining testbeds, focusing on hardware, software, and data. Developing Roadmap: Bring together experts to examine recent advances, identify current challenges, and develop a feasible roadmap to address them. Encouraging Engagement: Encourage more engagement in CAVs and robot swarms.

Attendees will gain insights into recent advances in the field and contribute to a summary of the workshop's discussions and findings that will be published on our website.

Keywords: Simulation and Real-World Testing Methodologies, Automated Vehicles, Vehicle Control and Motion Planning
Abstract: This paper presents the design of a research platform for autonomous driving applications, the Delft’s Autonomous-driving Robotic Testbed (DART). Our goal was to design a small-scale car-like robot equipped with all the hardware needed for on-board navigation and control while keeping it cost-effective and easy to replicate. To develop DART, we built on an existing off-the-shelf model and augmented its sensor suite to improve its capabilities for control and motion planning tasks. We detail the hardware setup and the system identification challenges to derive the vehicle's models. Furthermore, we present some use cases where we used DART to test different motion planning applications to show the versatility of the platform. Finally, we provide a git repository with all the details to replicate DART, complete with a simulation environment and the data used for system identification.

Keywords: Simulation and Real-World Testing Methodologies, Integration of Infrastructure and Intelligent Vehicles, Perception Including Object Event Detection and Response (OEDR)
Abstract: In recent years, small-scale driving testbeds have been developed as controlled physical environments for the evaluation of autonomous vehicle controllers. Such controllers are heavily dependent on computer vision algorithms that enable the vehicle to perceive its surroundings. To bridge the Sim2Real content and appearance gap between simulated and real-world image data for training these algorithms, we propose a novel transfer learning approach that performs domain adaptation using StyleGAN to generate style-mixed images that closely resemble real-world images. We explain our approach within the context of our small-scale driving testbed, CHARTOPOLIS, and demonstrate it on synthetic image data of two object classes, vehicles and buildings, from the driving simulator CARLA. Our results show that this approach works on the vehicle object class while failing on the building object class. This paper thus provides a plausible approach to bridging the Sim2Real gap through the use of custom pipelines that augment image datasets using a mix of techniques for domain adaptation and domain randomization.

Keywords: Perception Including Object Event Detection and Response (OEDR), Sensor Fusion for Localization, Cooperative Vehicles
Abstract: In this paper, we present an innovative indoor positioning system designed for computing the position and orientation of multiple model-scale vehicles. Our system, equipped with four cameras, utilizes a novel image-stitching technique to stitch the images from these cameras into a single output image. We explore diverse methodologies for detecting and estimating the position and orientation of the model-scale vehicles, demonstrating robust detection within the plane. Through rigorous testing with vehicles of varying dimensions, our system achieves an accuracy ranging from approximately 0.5cm to 3cm on both axes. This establishes the system's potential applicability in scenarios involving model-scale autonomous vehicles, where precise knowledge of each vehicle's position is critical.

Keywords: Vehicle-To-Everything (V2X) and Cellular V2X (C-V2X) Communications, Simulation and Real-World Testing Methodologies, Cooperative Vehicles
Abstract: In this work, we present a small-scale testbed for evaluating the real-life performance of cellular V2X (C-V2X) applications on 5G cellular networks. Despite the growing interest and rapid technology development for V2X applications, researchers still struggle to prototype V2X applications with real wireless networks, hardware, and software in the loop in a controlled environment. To help alleviate this challenge, we present a testbed designed to accelerate development and evaluation of C-V2X applications on 5G cellular networks. By including a small-scale vehicle platform into the testbed design, we significantly reduce the time and effort required to test new C-V2X applications on 5G cellular networks. With a focus around the integration of small-scale vehicle platforms, we detail the design decisions behind the full software and hardware setup of commonly needed intelligent transport system agents (e.g. sensors, servers, vehicles). Moreover, to showcase the testbed's capability to produce industrially-relevant, real world performance evaluations, we present an evaluation of a simple test case inspired from shared situational awareness. Finally, we discuss the upcoming use of the testbed for evaluating 5G cellular network-based shared situational awareness and other C-V2X applications.

Keywords: Automated Vehicles
Abstract: Interaction-driven behavior prediction and planning for autonomous vehicles

https://saschahornauer.github.io/

Topics

The topics of interest of the workshop include, but are not limited to:

Cooperative and comprehensible motion planning Probabilistic decision making and motion planning (including MDPs, POMDPs, MMDPs) Probabilistic behavior prediction (with help of semantic high-definition maps)

Second-order effects in heavy interactive scenarios Evaluation and benchmarking of the aforementioned topics

Keywords: End-To-End (E2E) Autonomous Driving, Vehicle Control and Motion Planning, Automated Vehicles
Abstract: Reinforcement learning has emerged as an important approach for autonomous driving. A reward function is used in reinforcement learning to establish the learned skill objectives and guide the agent toward the optimal policy. Since autonomous driving is a complex domain with partly conflicting objectives with varying degrees of priority, developing a suitable reward function represents a fundamental challenge. This paper aims to highlight the gap in such function design by assessing different proposed formulations in the literature and dividing individual objectives into Safety, Comfort, Progress, and Traffic Rules compliance categories. Additionally, the limitations of the reviewed reward functions are discussed, such as objectives aggregation and indifference to driving context. Furthermore, the reward categories are frequently inadequately formulated and lack standardization. This paper concludes by proposing future research that potentially addresses the observed shortcomings in rewards, including a reward validation framework and structured rewards that are context-aware and able to resolve conflicts.

Keywords: Vehicle Control and Motion Planning, Advanced Driver Assistance Systems (ADAS), Automated Vehicles
Abstract: The current adaptive cruise control (ACC) systems are susceptible to disruptive behaviors such as mandatory cut-ins, commonly known as "road bullying". To address this issue, this paper introduces an anti-bullying adaptive cruise control (AACC) approach endowed with proactive right-of-way protection capabilities. It bears the following features: i) with the capability of preventing bullying from mandatory cut-ins; ii) with real-time field implementation capability. The proposed approach utilizes inverse optimal control (IOC) technology to discern the driving styles of other road users online, subsequently employing Stackelberg competition for motion planning. Extensive simulation results demonstrate the efficacy of the proposed approach in thwarting bullying from mandatory cut-ins. Additionally, the approach exhibits support for real-time field deployment by maintaining computation times of less than 50 milliseconds.

Keywords: Vehicle Control and Motion Planning, Perception Including Object Event Detection and Response (OEDR)
Abstract: In autonomous driving, accurately interpreting the movements of other road users and leveraging this knowledge to forecast future trajectories is crucial. This is typically achieved through the integration of map data and tracked trajectories of various agents. Numerous methodologies combine this information into a singular embedding for each agent, which is then utilized to predict future behavior. However, these approaches have a notable drawback in that they may lose exact location information during the encoding process. The encoding still includes general map information. However, the generation of valid and consistent trajectories is not guaranteed. This can cause the predicted trajectories to stray from the actual lanes. This paper introduces a new refinement module designed to project the predicted trajectories back onto the actual map, rectifying these discrepancies and leading towards more consistent predictions. This versatile module can be readily incorporated into a wide range of architectures. Additionally, we propose a novel scene encoder that handles all relations between agents and their environment in a single unified heterogeneous graph attention network. By analyzing the attention values on the different edges in this graph, we can gain unique insights into the neural network's inner workings leading towards a more explainable prediction.

Keywords: Verification and Validation Techniques
Abstract: The accurate prediction of surrounding traffic actors’ movements is vital for the large-scale safe deployment of autonomous vehicles. Existing motion forecasting methods primarily aim to minimize prediction error by optimizing a loss function, which can sometimes lead to physically infeasible predictions or predictions that violate external constraints. This paper proposes a method that integrates explicit knowledge priors, allowing a network to forecast future trajectories that comply with both the vehicle’s kinematic constraints and the driving environment’s geometry. This is achieved by introducing a non-parametric pruning layer and attention layers to incorporate the defined knowledge priors. The proposed method aims to ensure reachability guarantees for traffic actors in both complex and dynamic situations. By conditioning the network to adhere to physical laws, we can achieve accurate and safe predictions, which are crucial for maintaining the safety and efficiency of autonomous vehicles in real-world settings.

Keywords: Automated Vehicles, Human Factors for Intelligent Vehicles
Abstract: Predicting the future behavior of human road users is an important aspect for the development of risk-aware autonomous vehicles. While many models have been developed towards this end, effectively capturing and predicting the variability inherent to human behavior still remains an open challenge. This paper proposes TrajFlow - a new approach for probabilistic trajectory prediction based on Normalizing Flows. We reformulate the problem of capturing distributions over trajectories into capturing distributions over abstracted trajectory features using an autoencoder, simplifying the learning task of the Normalizing Flows. TrajFlow outperforms state-of-the-art behavior prediction models in capturing full trajectory distributions in two synthetic benchmarks with known true distributions, and is competitive on the naturalistic datasets ETH/UCY, rounD, and nuScenes. Our results demonstrate the effectiveness of TrajFlow in probabilistic prediction of human behavior.

Keywords: Automated Vehicles
Abstract: Workshop on Off-Road Autonomy

https://sites.google.com/view/ivws-offroadautonomy/

Invited Speakers

Joel Pazhayapalli, CEO, Bluespace, "Field-testing Results on a Motion-based Odometry and Perception Stacks"

Fu Zhang, Prof., Univ. Hong Kong, "LiDAR-based Autonomous UAVs"

Michael Milford, Prof., Queensland Univ. Technology (QUT), "Introspection, Localization and Terrain Detection for Off-Road Autonomous Vehicles in Mine Sites and All Terrain Environments"

Topic of Interest

This workshop solicits high-quality technical papers. The topics of interest include but not limited to the following:

Sensor fusion for estimating traversability

Data set on and for off-road driving

Exploratory maneuvers

Calibration methods for easier and quicker transfer of autonomy stack

Transfer learning for applying urban and/or on-road AD stack to off-road

Economic learning approach for off-road maneuvers, e.g., online learning, self-supervised learning, etc.

Knowledge representation: Terrain, moving and static objects, weather

Long-term localization and mapping in off-road environments

Addressing the characteristics of off-road terrains, e.g., estimating lateral slips and mitigating them

Autonomous navigation for GPS-denied and rough terrains

State estimation / SLAM for off-road environments

Keywords: Sensor Signal Processing, Sensor Fusion for Localization, Automated Vehicles
Abstract: Autonomous off-road navigation requires an accurate semantic understanding of the environment, often converted into a bird's-eye view (BEV) representation for various downstream tasks. While learning-based methods have shown success in generating local semantic terrain maps directly from sensor data, their efficacy in off-road environments is hindered by challenges in accurately representing uncertain terrain features. This paper presents a learning-based fusion method for generating dense terrain classification maps in BEV. By performing LiDAR-image fusion at multiple scales, our approach enhances the accuracy of semantic maps generated from an RGB image and a single-sweep LiDAR scan. Utilizing uncertainty-aware pseudo-labels further enhances the network's ability to learn reliably in off-road environments without requiring precise 3D annotations. By conducting thorough experiments using off-road driving datasets, we demonstrate that our method can improve accuracy in off-road terrains, validating its efficacy in facilitating reliable and safe autonomous navigation in challenging off-road settings.

Keywords: Sensor Signal Processing, SLAM (Simultaneous Localization and Mapping), Automated Vehicles
Abstract: Recognizing traversable terrain from 3D point cloud data is critical, as it directly impacts the performance of autonomous navigation in off-road environments. However, existing segmentation algorithms often struggle with challenges related to changes in data distribution, environmental specificity, and sensor variations. Moreover, when encountering sunken areas, their performance is frequently compromised, and they may even fail to recognize them. To address these challenges, we introduce B-TMS, a novel approach that performs map-wise terrain modeling and segmentation by utilizing Bayesian generalized kernel (BGK) within the graph structure known as the tri-grid field (TGF). Our experiments encompass various data distributions, ranging from single scans to partial maps, utilizing both public datasets representing urban scenes and off-road environments, and our own dataset acquired from extremely bumpy terrains. Our results demonstrate notable contributions, particularly in terms of robustness to data distribution variations, adaptability to diverse environmental conditions, and resilience against the challenges associated with parameter changes.

Keywords: Vehicle Control and Motion Planning
Abstract: In the field of off-road navigation, integrating terrain data into path planning is becoming increasingly vital. It considers terrain roughness, slope, and step height, which are crucial parameters to ensure stability. This approach significantly differs from indoor driving scenarios, where the terrain is generally flat and exhibits less variability. In this paper, we define the traversability of terrain, numerically quantify it in terms of cost, and apply various asymptotically optimal planners, including RRT-Connect, RRT*, and PRM*, to identify the most cost-effective optimally traversable path. These planners are specifically designed to iteratively improve their solutions over time, gradually approaching the optimal solution as computation time increases. The effectiveness of these planners is evaluated within a limited time budget to assess their performance in simulation under realistic off-road conditions.

Keywords: Vehicle Control and Motion Planning, Automated Vehicles, Advanced Driver Assistance Systems (ADAS)
Abstract: This paper presents a real-time optimal trajectory generator and a robust controller designed to ensure the stability and performance of an unmanned off-road vehicle. The two-degree-of-freedom controller aims to achieve predictive driving control when perceptual information is informative and correct, while simultaneously attaining robustness through a fast feedback controller. For the feedback controller, we employ a recently developed approach called System Level Synthesis, providing transparency for robustness analysis and convexity for numerical optimization. At the same time, we focus on realizing the proposed algorithm as a practical means of online trajectory generation and control. The robust performance is investigated through extensive numerical simulations and experimental tests for an unmanned vehicle successfully executing navigation missions in an uncertain environment.

Keywords: Sensor Fusion for Localization, Sensor Signal Processing, Automated Vehicles
Abstract: With the rapid development of autonomous driving and SLAM technology, the performance of autonomous systems using multimodal sensors highly relies on accurate extrinsic calibration. Addressing the need for a convenient, maintenance-friendly calibration process in any natural environment, this paper introduces Galibr, a fully automatic targetless LiDAR-camera extrinsic calibration tool designed for ground vehicle platforms in any natural setting. The method utilizes the ground planes and edge information from both LiDAR and camera inputs, streamlining the calibration process. It encompasses two main steps: an initial pose estimation algorithm based on ground planes (GP-init), and a refinement phase through edge extraction and matching. Our approach significantly enhances calibration performance, primarily attributed to our novel initial pose estimation method, as demonstrated in unstructured natural environments, including on the KITTI dataset and the KAIST quadruped dataset.

Keywords: Automated Vehicles
Abstract: Autoware – ROS-based OSS for Autonomous Driving

https://autoware.org/iv2024/

Autoware is the world’s first and largest open-source project and community around software and hardware for self-driving. Autoware is based on ROS (Robot Operating System) middleware, with all the necessary features required for fully autonomous driving: sensing, localization, perception, planning, and control; it supports multiple vehicle interfaces as well as map formats.

The Autoware Foundation (AWF) has over 70 members from academia, industry, and government, contributing to the shared resources to cover the full spectrum of software and hardware for the autonomous driving ecosystem. Autoware has found widespread adoption: it is used by hundreds of companies, runs on 30+ vehicle types, and is used in 20+ countries. Autoware increasing development includes Automated Valet Parking (AVP) in 2020, Cargo Delivery in 2021, RoboBus and Racing vehicles for 2022. Newer projects include the Open AD Kit (demonstrated at IROS 2023) and the MIH Open EV platform.

Keywords: Simulation and Real-World Testing Methodologies, Functional Safety in Intelligent Vehicles, Automated Vehicles
Abstract: Extensive testing is necessary to ensure the safety of autonomous driving modules. In addition to component tests, the safety assessment of individual modules also requires a holistic view at system level, which can be carried out efficiently with the help of simulation. Achieving seamless compatibility between a modular software stack and simulation is complex and poses a significant challenge for many researchers. To ensure testing at the system level with state-of-the-art AV software and simulation software, we have developed and analyzed a bridge connecting the CARLA simulator with the AV software Autoware Core/Universe. This publicly available bridge enables researchers to easily test their modules within the overall software. Our investigations show that an efficient and reliable communication system has been established. We provide the simulation bridge as open-source software at https://github.com/TUMFTM/Carla-Autoware-Bridge

Keywords: Simulation and Real-World Testing Methodologies, Vehicle Control and Motion Planning, Verification and Validation Techniques
Abstract: Simulation is crucial in the development of autonomous driving software. In particular, assessing control algorithms requires an accurate vehicle dynamics simulation. However, recent publications use models with varying levels of detail. This disparity makes it difficult to compare individual control algorithms. Therefore, this paper aims to investigate the influence of the fidelity of vehicle dynamics modeling on the closed-loop behavior of trajectory-following controllers. For this purpose, we introduce a comprehensive Autoware-compatible vehicle model. By simplifying this, we derive models with varying fidelity. Evaluating over 550 simulation runs allows us to quantify each model's approximation quality compared to real-world data. Furthermore, we investigate whether the influence of model simplifications changes with varying margins to the acceleration limit of the vehicle. From this, we deduce to which degree a vehicle model can be simplified to evaluate control algorithms depending on the specific application. The real-world data used to validate the simulation environment originate from the Indy Autonomous Challenge race at the Autodromo Nazionale di Monza in June 2023. They show the fastest fully autonomous lap of TUM Autonomous Motorsport, with vehicle speeds reaching 267 km/h and lateral accelerations of up to 15 m/s^2.

Keywords: Simulation and Real-World Testing Methodologies, Verification and Validation Techniques
Abstract: Reproducibility is fundamental to robotics research, but access to appropriate hardware is a major limiting factor of reproducing experiments. This limitation usually arises from custom-built solutions without sufficient documentation and, simply, high costs. We propose that the best way to improve reproducibility is by removing barriers to safely sharing hardware. Towards this, we present a framework for making experimental environments remotely accessible. Modern virtualization tools like Linux containers are leveraged to enable reproducible and isolated access. The basic idea is to run user code in a container and transport all inputs and outputs through proxy programs that monitor for unsafe states. This interface is low-level: proxies can operate across serial lines, TCP, UDP, or HTTP connections. Thus, users are not locked into any particular programming library or application-level messaging system. The framework is shown in four case studies: the social robot Misty, the mobile robot Kobuki (base of the popular TurtleBot 2) with a LiDAR, wireless sensor networks, and Autoware.

Keywords: Vehicle Control and Motion Planning, Automated Vehicles
Abstract: The complex mathematical model of autonomous vehicles makes it difficult for system identification due to a combination of non-linearity and uncertainty. Various strategies have been proposed to address the difficulty in system identification, as it significantly influences the precise path-tracking performance of autonomous vehicles. This paper proposes a Koopman operator approach data-driven optimal control algorithm for path-tracking of autonomous vehicles. To identify mathematical model's various vehicle types of autonomous vehicle driving data were acquired in virtual simulation and real-world environments. An integrated linear model was identified using the Koopman operator neural network and the acquired driving data of autonomous vehicles. The identified integrated linear model was incorporated into a model predictive control algorithm designed for the path-tracking of autonomous vehicles. Reasonable path tracking performance was confirmed through performance evaluations conducted in path-tracking scenarios using various vehicle types for real and virtual vehicles in the real autonomous driving playground C-track and CARLA simulator environments.

Keywords: Vehicle Control and Motion Planning, Automated Vehicles, Simulation and Real-World Testing Methodologies
Abstract: This paper proposes a control technique for autonomous RC car racing. The presented method does not require any map-building phase beforehand since it operates only local path planning on the actual LiDAR point cloud. Racing control algorithms must have the capability to be optimized to the actual track layout for minimization of lap time. In the examined one, it is guaranteed with the improvement of the Stanley controller with additive control components to stabilize the movement in both low and high-speed ranges, and with the integration of an adaptive lookahead point to induce sharp and dynamic cornering for traveled distance reduction. The developed method is tested on a 1/10-sized RC car, and the tuning procedure from a base solution to the optimal setting in a real F1Tenth race is presented. Furthermore, the proposed method is evaluated with a comparison to a more simple reactive method, and in parallel to a more complex optimization-based technique that involves offline map building the global optimal trajectory calculation. The performance of the proposed method compared to the latter, referring to the lap time, is that the proposed one has only 8% lower average speed. This demonstrates that with appropriate tuning, a local planning-based method can be comparable with a more complex optimization-based one. Thus, the performance gap is lower than 10% from the state-of-the-art method. Moreover, the proposed technique has significantly higher similarity to real scenarios, therefore the results can be interesting in the context of automotive industry.

Keywords: End-To-End (E2E) Autonomous Driving, Automated Vehicles, Vehicle Control and Motion Planning
Abstract: This paper explores the use of reinforcement learning (RL) in the context of autonomous vehicle racing, specifically focusing on the F1TENTH simulation platform. While commercial autonomous driving often employs classic control algorithms, the state-of-the-art solutions, including those in the F1TENTH domain, increasingly rely on RL. Notably, RL-based approaches have shown superhuman performance in simulated environments, as seen in drone racing and the recent achievement by Sony in autonomous racing. In this paper we propose a novel LiDAR-only observation for learning vehicle dynamics, and test it with a widely accessible model-free RL method. The trained agent demonstrates the capability to transfer its driving skills to previously unseen tracks. Additionally, the paper provides recommendations for selecting hyperparameters, contributing valuable insights for newcomers to the field of autonomous racing.

Keywords: End-To-End (E2E) Autonomous Driving, Software-Defined Vehicle for Intelligent Vehicles, Automated Vehicles
Abstract: In recent years, the number and importance of autonomous racing leagues, and consequently the number of studies on them, has been growing. The seamless integration between different series has gained attention due to the scene's diversity. However, the high cost of full scale racing makes it a more accessible development model, to research at smaller form factors and scale up the achieved results. This paper presents a scalable architecture designed for autonomous racing that emphasizes modularity, adaptability to diverse configurations, and the ability to supervise parallel execution of pipelines that allows the use of different dynamic strategies. The system showcased consistent racing performance across different environments, demonstrated through successful participation in two relevant competitions. The results confirm the architecture's scalability and versatility, providing a robust foundation for the development of competitive autonomous racing systems. The successful application in real-world scenarios validates its practical effectiveness and highlights its potential for future advancements in autonomous racing technology.

Keywords: Integration of HD map and Onboard Sensors, Sensor Fusion for Localization, Verification and Validation Techniques
Abstract: Today’s software stacks for autonomous vehicles rely on HD maps to enable sufficient localization, accurate path planning, and reliable motion prediction. Recent developments have resulted in pipelines for the automated generation of HD maps to reduce manual efforts for creating and updating these HD maps. We present FlexMap Fusion, a methodology to automatically update and enhance existing HD vector maps using OpenStreetMap. Our approach is designed to enable the use of HD maps created from LiDAR and camera data within Autoware. The pipeline provides different functionalities: It provides the possibility to georeference both the point cloud map and the vector map using an RTK-corrected GNSS signal. Moreover, missing semantic attributes can be conflated from OpenStreetMap into the vector map. Differences between the HD map and OpenStreetMap are visualized for manual refinement by the user. In general, our findings indicate that our approach leads to reduced human labor during HD map generation, increases the scalability of the mapping pipeline, and improves the completeness and usability of the maps. The methodological choices may have resulted in limitations that arise especially at complex street structures, e.g., traffic islands. Therefore, more research is necessary to create efficient preprocessing algorithms and advancements in the dynamic adjustment of matching parameters. In order to build upon our work, our source code is available at https://github.com/TUMFTM/FlexMap_Fusion.

Keywords: Sensor Signal Processing, Sensor Fusion for Localization
Abstract: Nowadays, using LiDARs, RGB cameras and Thermal cameras for automatic systems, in particular self-driving cars, has become the common approach in multiple deployments. Each kind of sensor has distinct advantages, leading to the fact that using multiple sensors can help autonomous systems improve their performance. Calibration between sensors is the precondition of fusing multiple sensors. This paper presents a novel way to register extrinsic parameters for LiDAR, 360 RGB camera and 360 Thermal camera automatically based on features information. To evaluate the method, we use our dataset around Nagoya University.

Keywords: Automated Vehicles
Abstract: The 12th Workshop and Industry Panel on Cooperative and Automated Driving

Sponsors and Co-Organisers: IEEE Intelligent Transportation Systems Society (IEEE ITSS) IEEE Future Networks Technical Community (IEEE FNTC) IEEE Standards Association (IEEE SA) Program:

13:00-13:10 Dr. Meng Lu (IEEE ITSS & IEEE FNTC): Openging and Introduction

13:10-13:30 Kaj Munhoz Arfvidsson, Frank J. Jiang*, Prof. Karl H. Johansson, Prof. Jonas Mårtensson (KTH Royal Institute of Technology, Sweden): Ensuring Safety at Intelligent Intersections: Temporal Logic Meets Reachability Analysis.

13:30-13:50 Hao Su, Prof. Shin'ichi Arakawa*, Prof. Masayuki Murata (Osaka University, Japan): Cooperative 3D Multi-Object Tracking with Complementary Data Association for Connected and Automated Vehicles

13:50-14:05 Dr. Werner Ritter (Mercedes-Benz AG, Germany): AI-SEE: Artificial Intelligence Enhancing Vehicle Vision in Low Visibility Conditions – Overview and Latest Results

14:05-14:20 Prof. Alwin Kienle (Institute for Laser Technologies in Medicine and Metrology, University of Ulm, Germany): Physics-based Rendering of Inclement Weather

14:20-14:35 Dr. Matti Kutila (VTT Technical Research Centre of Finland Ltd., Finland): Automated Driving in Harsh Winter Weather

14:35-14:45 Break

14:45-15:00 Rob Gee (Continental, USA): On the Way to the Tipping Point: Complexities for Deployment of Connected and Intelligent Vehicle Systems.

15:00-15:15 Anderson Ferraz (Denso, Germany): CONNECT: Continuos and Dynamic Trust Assessment for Cooperative Mobility

15:15-15:30 Dr. Sergei Avedisov (Toyota North America R&D, USA): Enhancing Localization, Perception and Maneuvering through Connectivity.

15:30-15:45 Vandana Narri (Scania, Sweden): Enhancement of Autonomous Driving Platform using Data Sharing Between Infrastructure and Vehicle

15:45-16:20 CAD Industry Panel (Panelists: Invited Speakers and IEEE ITSS/SC Members)

Keywords: Vehicle-To-Everything (V2X) and Cellular V2X (C-V2X) Communications, Sensor Signal Processing, Cooperative Vehicles
Abstract: Cooperative perception has attracted sustained attention, promising groundbreaking contributions to transportation safety and efficiency. It enables vehicles to share environmental information in addressing limited visibility, thus improving individual perception performance. However, most related studies only focus on detection, and ways to explicitly enhance object tracking capabilities through multi-vehicle cooperation still lack sufficient exploration. In this paper, we propose a cooperative 3D multi-object tracking (MOT) system that leverages complementary information from multiple vehicles to alleviate the problem of temporary tracking failures. Specifically, we design a data association module to assist the ego vehicle in leveraging received information to promptly compensate for its missed objects. To avoid erroneous associations, we maintain an object ID mapping set for each communication link to discover the correspondence between objects tracked by different vehicles. We conduct experiments on the V2V4Real dataset and utilize the official pre-trained network checkpoints to generate detection candidates as inputs. Experimental results demonstrate that the proposed method performs favorably against the baseline without bringing a communication burden, as well as its generalizability for various detectors.

Keywords: Smart Infrastructure, Cooperative Vehicles, Automated Vehicles
Abstract: In this work, we propose an approach for ensuring the safety of vehicles passing through an intelligent intersection. There are many proposals for the design of intelligent intersections that introduce central decision-makers to intersections for enhancing the efficiency and safety of the vehicles. To guarantee the safety of such designs, we develop a safety framework for intersections based on temporal logic and reachability analysis. We start by specifying the required behavior for all the vehicles that need to pass through the intersection as linear temporal logic formula. Then, using temporal logic trees, we break down the linear temporal logic specification into a series of Hamilton-Jacobi reachability analyses in an automated fashion. By successfully constructing the temporal logic tree through reachability analysis, we verify the feasibility of the intersection specification. By taking this approach, we enable a safety framework that is able to automatically provide safety guarantees on new intersection behavior specifications. To evaluate our approach, we implement the framework on a simulated T-intersection, where we show that we can check and guarantee the safety of vehicles with potentially conflicting paths.

Keywords: Automated Vehicles
Abstract: t

Fourth Workshop on Intelligent Transportation Systems, Intelligent Vehicles and Advanced Driver Assistant Systems for Unstructured Environments (ITSIVUE 2024) https://sites.google.com/view/itsivue2024/

There is a growing demand for Intelligent Transportation Systems (ITS), Intelligent Vehicles (IV), Advanced Driver Assistant Systems (ADAS) and Assistive Mobility (AM) in unstructured environments. Current solutions deployed by vehicle manufacturers in the ITS, IV and ADAS space are based on the assumptions that the environment is well-structured with well-defined lanes, road surface markings, traffic signs and lights and stringent and well-defined traffic rules are followed.

Keywords: Advanced Driver Assistance Systems (ADAS), Policy, Ethics, and Regulations, Automated Vehicles
Abstract: The aim of this paper is to investigate the relationship between operational design domains (ODD), autonomous driving SAE Levels, and Technology readiness level (TRL). The first autonomous vehicles, like robotaxis, are in commercial use, and the first vehicles with highway pilot systems have been delivered to private customers. % Level 4 automated valet parking is being tested in the first parking garages.It has emerged as a crucial issue that those autonomous driving systems differ significantly in their ODD and in their technical maturity. Consequently, any approach to compare those systems is difficult and requires a deep dive into defined ODDs, specifications, and used technologies. Therefore, this paper challenges current state-of-the-art taxonomies and develops a new and integrated taxonomy that can structure automated vehicle systems more efficiently. We use the well-known SAE Levels 0-5 as the "level of responsibility", and connect and describe the ODD on an intermediate level of abstraction. This method was then used to analyze today's existing autonomous vehicle applications, which are structured into the new taxonomy and rated by the new maturity levels. Finally, a new maturity model is explicitly proposed to improve the comparability of automated vehicles and driving functions. Our results indicate that this new taxonomy and maturity level model will help to differentiate autonomous vehicle systems in discussions more clearly and to discover white fields more systematically and upfront, e.g., for research but also for regulatory purposes.

Keywords: Advanced Driver Assistance Systems (ADAS), Automated Vehicles, Perception Including Object Event Detection and Response (OEDR)
Abstract: Advanced Driver Assistance Systems (ADAS) are rapidly becoming a standard feature in modern road vehicles, enhancing safety and driver comfort. As ADAS adoption expands across diverse geographical and cultural regions, the performance of camera-based perception systems may vary sig- nificantly due to environmental and expected social behaviour of the different actors. This paper explores the referred factors and evaluates the traffic environment complexity for vehicles with different levels of automation. In particular, we propose a novel modeling and quantitative assessment approach for environment complexity. Specifically, we compare a perception model trained on United States dataset with a dataset from India, a nation characterized by unique traffic patterns, signage conventions, and cultural norms to assess its performance variation, and to lay the basis for proposing influencing factors of traffic environment complexity. We establish a scheme of referential and additional static factors and based on an expert evaluation, environment complexity is established. The effectiveness of the proposed approach is testified by naturalistic driving data. These findings pave the way for future research in intelligent driving and emphasize the importance of addressing cultural nuances as vehicle automation levels increase

Keywords: Pedestrian Protection, Perception Including Object Event Detection and Response (OEDR), Human Factors for Intelligent Vehicles
Abstract: Developing automatic mobility assistance systems for the safe navigation of visually impaired pedestrians in unstructured environments presents several complex challenges. The dynamic and unpredictable conditions and the diverse range of obstacles encountered by a pedestrian walking on an unstructured road add to the difficulties of automatic navigation. Another significant aspect of safe mobility assistance is real-time performance, which demands lightweight architectures and fast processing. In this paper, we propose an obstacle detection framework combining a road object detection model and a road anomaly detection model, using parallel processing for fast real-time performance. The models are based on convolutional neural network backbones, use transfer learning, and are trained on custom datasets manually collected in unstructured environments. The proposed system addresses the challenges of the complexities of the camera-based automatic navigation to detect obstacles and, based on their position and size, alerts the user via audio feedback.

Keywords: Automated Vehicles
Abstract: The 2nd International Workshop on Socially Interactive Autonomous Mobility (SIAM)

https://interactive-driving.github.io/

One of the main goals of our workshop is to bridge the gap between Computational Cognitive & Behavior Science, Explainable AI, Transportation, and the Autonomous Driving community. Our SIAM workshop mainly targets theoretical frameworks and practical algorithms of perception, decision-making, and planning integrated with social factors and computational cognitive science to enable autonomous vehicles (AVs) to interact with human agents in a socially compatible way. Specifically, the topics are as follows, but not limited to:

Applications of AVs interacting with human agents; Algorithms of perception, decision-making, planning for human-like AVs;

Cognitive aspects and models for autonomous driving;

Cognitive and mental modeling toward socially driving, e.g., Theory of Mind and Theory of Machine;

Social cues for AVs in interactive driving tasks;

Action-reaction cycle modeling and validation;

Explainable interaction and planning in interactive driving tasks;

Evaluation and quantification of inter-human interactions and their implementations to human-AV interactions;

Human driving behavior/intention modeling, simulation, and analysis;

Heterogeneous human-agent teams;

Interactive traffic scenes analysis;

Interaction pattern learning, extraction, and recognition;

Interactive simulations and humans-in-the-loop simulations;

Learning-based theory for social interaction among human drivers;

Social and group intelligence in multiple human agent interaction;

Spatiotemporal driving behaviors in interactive traffic scenes;

Keywords: Automated Vehicles, Automotive Cyber Physical Systems, Simulation and Real-World Testing Methodologies
Abstract: Automated vehicles (AVs) hold the potential to significantly improve traffic flow, reducing travel time, energy consumption, and emissions. However, until AVs achieve high market penetration rates, navigating the transition to mixed-autonomy traffic — comprising both AVs and human-driven vehicles (HVs) — presents substantial challenges. While numerous studies have concentrated on AV control within mixed-autonomy environments, human-AV interactions have been largely neglected. To understand the benefits of considering the impact of AVs on their followers in mixed traffic control, we introduce a general framework focused on social interaction-aware benefits. Through this framework, we develop an interaction-aware control approach aimed at optimizing socially compatible traffic flow. The results demonstrate that as social interactions between the AV and its following HVs are considered, the benefits for the AV may decrease. In contrast, HVs can gain more benefits when the interaction-aware control strategy is not solely focused on the AV.

Keywords: Advanced Driver Assistance Systems (ADAS), Functional Safety in Intelligent Vehicles, Automotive Cyber Physical Systems
Abstract: Automated Vehicles (AVs), particularly those with Adaptive Cruise Control (ACC), are increasingly integral to intelligent transportation systems, but they bring new cybersecurity challenges. This study explores the subtleties of cyberattacks targeting ACC vehicles, specifically through false data injection, and assesses their impact on traffic dynamics. We innovatively adapt and implement strategically designed cyberattacks in simulations, providing a realistic evaluation of their effects on traffic. Our approach not only synthesizes these attacks but also rigorously tests their detectability against state-of-the-art detection algorithms. The findings reveal the intrinsic difficulty in detecting such stealthily designed attacks, highlighting a significant gap in current cybersecurity measures. Despite the precision of our detection methods, the low recall rates emphasize the stealthiness of these attacks. The open-sourced experiment code is made available at https://github.com/tianyi17/simulations_IV24. This research accentuates the urgent need for more sophisticated detection and defense strategies to protect ACC vehicles against evolving cyber threats, ensuring the reliability and safety of future transportation systems.

Keywords: Automated Vehicles, Human Factors for Intelligent Vehicles, Vehicle Control and Motion Planning
Abstract: This research aims to investigate professional racing drivers’ expertise to develop an understanding of their cognitive and adaptive skills to create new autonomy algorithms. An expert interview study was conducted with 11 professional race drivers, data analysts, and racing instructors from across prominent racing leagues. The interviews were conducted using an exploratory, non-standardized expert interview format guided by a set of prepared questions. The study investigates drivers' exploration strategies to reach their vehicle limits and contrasts them with the capabilities of state-of-the-art autonomous racing software stacks. Participants were questioned about the techniques and skills they have developed to quickly approach and maneuver at the vehicle limit, ultimately minimizing lap times. The analysis of the interviews was grounded in Mayring's qualitative content analysis framework, which facilitated the organization of the data into multiple categories and subcategories. Our findings create insights into human behavior regarding reaching a vehicle’s limit and minimizing lap times. We conclude from the findings the development of new autonomy software modules that allow for more adaptive vehicle behavior. By emphasizing the distinct nuances between manual and autonomous driving techniques, the paper encourages further investigation into human drivers’ strategies to maximize their vehicles’ capabilities.

Keywords: Cooperative Vehicles, Vehicle Control and Motion Planning, Future Mobility and Smart City
Abstract: Autonomous driving is an advanced field that attracts significant attention and engages numerous researchers. However, relying solely on a single autonomous vehicle (AV) is insufficient to meet the demand of future transportation systems. This necessitates the application of connected and autonomous vehicles (CAVs), whose operation relies on multi-agent decision-making technology. Currently, research primarily focuses on simple traffic scenarios. However, unsignalized intersections are frequently encountered in rural areas, characterized by high traffic volume, complex interactions, and significant risks. It is crucial to conduct research on the decision-making of CAVs at unsignalized intersections. To address these issues, the lane-changing decision-making of large-scale AVs at unsignalized intersections is studied in this paper. First, an action filter-based deep Q-network method named AF-DQN is proposed, which enables AVs to effectively filter out potentially hazardous lane-changing actions and execute safe actions. Additionally, a multi-objective reward function that considers multiple factors has been designed, including safety, task achievement, and compliance. Moreover, an exploratory training strategy is introduced to train the multi-agent deep reinforcement learning network model. The strategy facilitates agents to learn through exploration in simple scenarios before solving complex driving tasks in more complex scenarios. Finally, experiments are conducted to validate the effectiveness and superiority of the proposed method. Results show that exploratory training accelerates the model's training speed and improves training effectiveness. Moreover, the AF-DQN method outperforms the baseline method in terms of safety, efficiency, and adherence to traffic rules.

Keywords: Human Factors for Intelligent Vehicles, Automated Vehicles
Abstract: This paper structures the human-intrinsic social interaction mechanism derived in the field of humanities and applied the designed mechanism into the driver-driver interaction in lane-merging scenarios. Social interaction mechanism consists of the intrinsic reasoning-based decision-making mechanism and negotiation mechanism. Negotiation mechanism for social interaction between drivers should be established both distributive strategy and integrative strategy because drivers sometimes adjust their decisions by yielding their utilities to maximize the mutual utility with the other driver. The structured social interaction mechanism is validated by comparing the extracted decision results with the real driving database with the simply designed utility. The proposed social interaction mechanism can be applied to explain the process and reason of decision-making for a lane-keeping and a lane-changing vehicle in lane-merging scenarios.

Keywords: Human Factors for Intelligent Vehicles, Verification and Validation Techniques, Cooperative Vehicles
Abstract: Effective driving style analysis is critical to developing human-centered intelligent driving systems that consider drivers' preferences. However, the approaches and conclusions of most related studies are diverse and inconsistent because no unified datasets tagged with driving styles exist as a reliable benchmark. The absence of explicit driving style labels makes verifying different approaches and algorithms difficult. This paper provides a new benchmark by constructing a Natural Dataset of Driving Style (NDDStyle) tagged with the subjective evaluation of 100 drivers' driving styles. In our dataset, the subjective quantification of each driver's driving style is from themselves and an expert according to the Likert-scale questionnaire. The testing routes are selected to cover various driving scenarios, including highways, urban, highway ramps, and signalized traffic. The collected driving data consists of lateral and longitudinal manipulation information collected from CAN, including steering angle, steering speed, lateral acceleration, throttle position, throttle rate, brake pressure, etc. This driving-style dataset is the first to provide detailed manipulation data with driving-style tags.

Keywords: Integration of HD map and Onboard Sensors, Future Mobility and Smart City, Pedestrian Protection
Abstract: This paper introduces RobotCycle, a novel ongoing project that leverages Autonomous Vehicle (AV) research to investigate how road infrastructure influences cyclist behaviour and safety during real-world journeys. The project’s requirements were defined in collaboration with key stakeholders, including city planners, cyclists, and policymakers, informing the design of risk and safety metrics and the data collection criteria. We propose a data-driven approach relying on a novel, rich dataset of diverse traffic scenes and scenarios captured using a custom-designed wearable sensing unit. By analysing road-user trajectories, we identify normal path deviations indicating potential risks or hazardous interactions related to infrastructure elements in the environment. Our analysis correlates driving profiles and trajectory patterns with local road segments, driving conditions, and road-user interactions to predict traffic behaviours and identify critical scenarios. Moreover, by leveraging advancements in AV research, the project generates detailed 3D High-Definition Maps (HD Maps), traffic flow patterns, and trajectory models to provide a comprehensive assessment and analysis of the behaviour of all traffic agents. These data can then inform the design of cyclist-friendly road infrastructure, ultimately enhancing road safety and cyclability. The project provides valuable insights for enhancing cyclist protection and advancing sustainable urban mobility.

Keywords: Pedestrian Protection, Verification and Validation Techniques, Simulation and Real-World Testing Methodologies
Abstract: Developing safer autonomous vehicles necessitates extensive testing of pedestrian behavior, particularly in atypical situations. Existing datasets lack consistent annotations, with text-based explanations and per-frame annotations causing redundancy and obscuring temporal relationships. To address these issues, we propose PedAnalyze, a Python-based annotator that focuses on pedestrian and vehicle behavior and facilitates structured datasets with pre-defined tags. Our approach allows for both single-frame and multi-frame annotations, which reduces the number of repetitive tasks. In addition, we focus on curating datasets from dash-cam videos on platforms such as YouTube, capturing valuable and rare pedestrian-vehicle incidents. We aim to create a comprehensive pedestrian behavior ontology and dataset to advance autonomous driving system research and development.

Keywords: Sensor Signal Processing, Perception Including Object Event Detection and Response (OEDR), Automated Vehicles
Abstract: As a vital component in autonomous driving, accurate trajectory prediction effectively prevents traffic accidents and improves driving efficiency. To capture complex spatial-temporal dynamics and social interactions, recent studies developed models based on advanced deep-learning methods. On the other hand, recent studies have explored the use of deep generative models to further account for trajectory uncertainties. However, the current approaches demonstrating indeterminacy involve inefficient and time-consuming practices such as sampling from trained models. To fill this gap, we proposed a novel model named Graph Recurrent Attentive Neural Process (GRANP) for vehicle trajectory prediction while efficiently quantifying prediction uncertainty. In particular, GRANP contains an encoder with deterministic and latent paths, and a decoder for prediction. The encoder, including stacked Graph Attention Networks, LSTM and 1D convolutional layers, is employed to extract spatial-temporal relationships. The decoder is used to learn a latent distribution and thus quantify prediction uncertainty. To reveal the effectiveness of our model, we evaluate the performance of GRANP on the highD dataset. Extensive experiments show that GRANP achieves state-of-the-art results and can efficiently quantify uncertainties. Additionally, we undertake an intuitive case study that showcases the interpretability of the proposed approach. The code is available at https://github.com/joy-driven/GRANP.

Keywords: Vehicle Control and Motion Planning, Cooperative Vehicles, Automated Vehicles
Abstract: The collaborative decision-making technology of connected and autonomous vehicles (CAVs) is critical in today's autonomous driving. Recently, graph reinforcement learning (GRL)-based methods have demonstrated exemplary performance in solving decision-making problems by implementing graphic technologies. However, current GRL-based research faces the challenge of modeling the interaction completely and extracting driving features efficiently. To address these issues, this paper proposes a social interaction force (SIF) spatial-temporal graph dynamic attention network (SIF-STGDAN) to solve the decision-making of CAVs. First, a SIF model is established to better represent the mutual effect between vehicles; an on-ramp merging scenario is then constructed and modeled by graph representation. Then, the SIF-STGDAN is proposed by combining the temporal convolutional network (TCN) and graph dynamic attention network to extract the graphic features of the on-ramp scenario efficiently, and the double deep q-learning (DDQN) algorithm is utilized to generate the optimized driving strategies for CAVs. Finally, experiments are conducted, and results show that our proposed SIF-STGDAN outperforms the baselines in terms of safety, efficiency, and model stability.

Keywords: Automated Vehicles
Abstract: Infrastructure Support and Impact in Autonomous Vehicle Deployment

https://sites.psu.edu/xbhu/ieee-intelligent-vehicle-symposi um-infrastructure-support-workshop/

Objectives: Our workshop aims to delve into the symbiotic relationship between autonomous vehicles (AVs) and the infrastructural environment. It is designed to foster discussions on the challenges and opportunities emerging from the unique demands AVs place on road infrastructure and the systemic adaptations required to support their deployment. We will explore novel research findings, engage in thought-provoking discussions, and provide networking opportunities for academics, industry professionals, and policymakers interested in the future of AV infrastructure.

Keywords: Integration of Infrastructure and Intelligent Vehicles, Automated Vehicles, Policy, Ethics, and Regulations
Abstract: Autonomous driving system is rapidly advancing, resulting in more Highly Automated Vehicles (HAVs) interacting with human-driven vehicles. In order to ensure safety during interaction, External Human-Machine Interface (EHMI) serves as a channel to display HAVs’ status and intentions. However, from the perspective of game theory, the selfish act of disclosing false information is detrimental to the advancement of HAV in most cases. It raises the question of whether certain acts of altruistic deception should be permitted if they can enhance the benefit for everyone in some certain cases. Therefore, it is worthwhile to investigate strategies for EHMI information disclosure. In this study, we establish a game-theoretic model of unprotected left-turn intersection scenarios. The impact of EHMI information disclosure is taken into consideration. The proposed model effectively achieves the simultaneous maximization of safety and efficiency during the human-machine interaction. In theory, altruistic deception may have benefits in certain scenarios, but from a practical standpoint, it is detrimental to the development of HAVs in the long run. Therefore, we recommend disclosing authentic information. By examining the benefits and potential drawbacks of altruistic deception, as well as considering the ethical considerations and legal implications, this study aims to contribute to the ongoing discourse on the ethical framework governing autonomous driving system.

Keywords: Vehicle-To-Everything (V2X) and Cellular V2X (C-V2X) Communications, Smart Infrastructure, Automated Vehicles
Abstract: Automated shuttles are currently limited to very simple and clear scenarios, e.g. with dedicated lanes, low traffic density and little flexibility in interaction with other road users. Furthermore, the vehicles are always monitored by a safety driver and the systems are highly adapted to a certain route and only suitable for one specific application. This whitepaper presents the general test site architecture, algorithmic challenges and a general overview of the German research project ”Shuttle2X”. It has the aim to go significantly beyond these dedicated solutions currently found in very isolated environments. This is achieved by expanding the operating area through research and development of new technologies, in particular through a connection to an intelligent infrastructure and selective expansion of automated driving (AD) capabilities as well as a highly reliable and func- tional safe communication with the infrastructure at dedicated and essential route points. For this purpose, functional algorithms are also being further developed using artificial intelligence (AI) processes by considering fault tolerance in order to enable robust self-driving operation and thus ultimately contribute to a replacement of the safety driver.

Keywords: Automated Vehicles
Abstract: Are You Happy with AV? User Experience (UX) in AV-Human Interaction in collaboration with the Workshop on Human Factors in Intelligent Vehicles

1) https://sites.google.com/view/are-you-happy-with-av-2024/pr ograms?authuser=0

2) http://hfiv.net/

1pm-1:05pm Opening and Introduction

1:05pm-1:25pm Invited speaker 1 Prof. Dr. Cristina Olaverri Monreal Johannes Kepler University Linz, Austria Enhancing Road Safety in Automated Driving: Exploring the Human Factor, Vulnerable Road Users, and Mixed Traffic Scenarios

1:25pm-1:45pm Invited speaker 2 Prof. Fang You & Prof. Jianmin Wang Tongji University, China Research on human-machine collaborative interaction design for intelligent driving

1:45pm-2:05pm Contributed Paper Masaki Kuge, Hailong Liu, Toshihiro Hiraoka, Takahiro Wada NAIST, Japan; JARI, Japan. An eHMI Presenting Request-To-Intervene Status of Level 3 Automated Vehicles to Surrounding Vehicles

2:05pm-2:25pm Workshop paper 1 Jemin Woo, Changsun Ahn Pusan National University, Republic of Korea Towards Human-Like Autonomous Vehicles: A Qualitative Evaluation and Design Perspective

2:25pm-2:40pm Break

2:40pm - 3pm Invited speaker 3 Prof. Dr. Takahiro Wada NAIST, Japan TBD

3pm - 3:20pm Invited speaker 4 Dr. Ignacio Alvarez Intel Lab, U.S.A. The use of Generative AI and Large Language models for UX Design in Automotive Applications – An introduction & tutorial

3:20pm - 3:40pm Clare Mutzenich, Fergus McVey, Ceire Martin, Claire Harding 7th Sense Research, United Kingdom Driving the Future: Addressing Generational Trust and Ownership Barriers in the Adoption of Connected and Autonomous Vehicles

3:40pm - 4pm Lucas Elbert Suryana, Sina Nordhoff, Simeon Craig Calvert, Arkady Zgonnikov, Bart van Arem Delft University of Technology, The Netherlands A Meaningful Human Control Perspective on User Perception of Partially Automated Driving Systems: A Case Study of Tesla Users

Keywords: Automated Vehicles, Vehicular Active and Passive Safety, Integration of Infrastructure and Intelligent Vehicles
Abstract: The advent of Connected and Autonomous Vehicles (CAVs) promises a transformative shift in transport, ushering in an era of shared mobility and interconnectedness. However, recent events highlight the challenges facing the widespread acceptance of CAVs and Mobility as a Service (MaaS). Drawing on insights from a survey of over 3,000 transport users in the UK, our study reveals two significant hurdles impeding the transition to shared autonomy. Firstly, user adoption presents a formidable challenge, slowing the pace of relinquishing control to driverless mobility, due to low trust and acceptance across generational groups. Secondly, the enduring preference for private vehicle ownership acts as a barrier to embracing shared mobility solutions. To address these challenges, we introduce the SASS Model, which categorises individuals into four distinct groups based on their emotional and rational inclinations towards CAVs and MaaS: Sceptics, Alarmists, Swing Voters, and Supporters. Each group necessitates a tailored approach to effectively communicate the benefits of CAVs. Strategies include addressing the concerns of Alarmists, providing reassurance to Sceptics, educating Swing Voters about the advantages of an automated future, and leveraging the backing of Supporters. Ultimately, the momentum of the automated (r)evolution will be driven by rational Sceptics and Swing Voters, highlighting the importance of targeted messaging and positioning by the automotive industry and transport providers to secure widespread adoption.

Keywords: Human Factors for Intelligent Vehicles, Automated Vehicles, Verification and Validation Techniques
Abstract: This study proposes a method for qualitatively evaluating and designing human-like driver models for autonomous vehicles. While most existing research on human-likeness has been focused on quantitative evaluation, it is crucial to consider qualitative measures to accurately capture human perception. To this end, we administered surveys to participants to discern whether the driver was human or autonomous. The survey employed direct experiential evaluation by participants. The findings of this research can significantly contribute to the development of naturalistic and human-like driver models for autonomous vehicles, enabling them to safely and efficiently coexist with human drivers in diverse driving scenarios through the interaction between human and autonomous vehicles.

Keywords: Automated Vehicles, Human Factors for Intelligent Vehicles, Policy, Ethics, and Regulations
Abstract: The use of partially automated driving systems raises concerns about potential responsibility issues, posing risk to the system safety, acceptance, and adoption of these technologies. The concept of meaningful human control has emerged in response to the responsibility gap problem, requiring the fulfillment of two conditions, tracking and tracing. While this concept has provided important philosophical and design insights on automated driving systems, there is currently little knowledge on how meaningful human control relates to subjective experiences of actual users of these systems. To address this gap, our study aimed to investigate the alignment between the degree of meaningful human control and drivers’ perceptions of safety and trust in a real-world partially automated driving system. We utilized previously collected data from interviews with Tesla ``Full Self-Driving'' (FSD) Beta users, investigating the alignment between the user perception and how well the system was tracking the users' reasons. We found that tracking of users' reasons for driving tasks (such as safe maneuvers) correlated with perceived safety and trust, albeit with notable exceptions. Surprisingly, failure to track lane changing and braking reasons was not necessarily associated with negative perceptions of safety. However, the failure of the system to track expected maneuvers in dangerous situations always resulted in low trust and perceived lack of safety. Overall, our analyses highlight alignment points but also possible discrepancies between perceived safety and trust on the one hand, and meaningful human control on the other hand. Our results can help the developers of automated driving technology to design systems under meaningful human control and are perceived as safe and trustworthy.

Keywords: Automated Vehicles
Abstract: Ensuring traffic safety and mitigating accidents in modern driving is of paramount importance, and computer vision technologies have the potential to significantly contribute to this goal. This paper presents a multi-modal Vision Transformer for Driver Distraction Detection (termed ViT-DD), which incorporates inductive information from training signals related to both distraction detection and driver emotion recognition. Additionally, a self-learning algorithm is developed, allowing for the seamless integration of driver data without emotion labels into the multi-task training process of ViT-DD. Experimental results reveal that the proposed ViT-DD surpasses existing state-of-the-art methods for driver distraction detection by 6.5% and 0.9% on the SFDDD and AUCDD datasets, respectively.

Keywords: Automated Vehicles
Abstract: Simulation provides a fast, cost-effective, and secure environment for developing autonomous driving systems. However, mitigating the gap between simulation and reality is a challenging task as it demands a behavior simulation method that is human-like, diverse, controllable, socially consistent, and scalable. This work proposes a data-driven traffic agent simulation method to address the aforementioned challenges. Our approach centers around a graph-based scene representation and an encoding method, dividing the simulation into two stages: Multi-Agent Goal assignment (MAG) and Goal-Directed Planning (GDP). Firstly, we create joint goal sets for all agents involved in the scenario. Subsequently, we assign target centerlines (TCLs) to each agent based on their predicted goals. To account for any potential mismatch between the predicted joint goal sets and the road structure, we further align the goals of each agent with their respective assigned TCLs. These on-TCL goals serve as inputs for our interactive autoregressive Goal-Directed Planner (AR-GDP), constituting the second stage of our method that generates roll-outs for simulations. Evaluation results on the leaderboard of the Waymo Open Sim Agents Challenge (WOSAC) 2023 show the competitiveness of the proposed method.

Keywords: Automated Vehicles
Abstract: In the field of Simultaneous Localization and Map Building (SLAM), robots have become highly proficient in self-localization. However, the localization and tracking of irregular objects in the environment still pose significant challenges. Consequently, this work proposes a real-time approach to detect moving objects in the environment and estimate their six-degree-of-freedom (6-DoF) motion without making any assumptions about the type of the objects. The central idea is to analyze the correlation between map points and segmenting point clouds which are parts of dynamic objects belonging to different groups. Then, a region-based bundle adjustment method has been developed to obtain the optimized pose of the objects on the SE(3) manifold. Our method surpasses existing appearance-based approaches, which struggle to handle irregular objects. To validate the effectiveness of our algorithm, we tested our algorithm in real-world environments. The results demonstrate our method achieves superior accuracy in tracking dynamic objects, showcasing its potential for various applications.

Keywords: Automated Vehicles
Abstract: This paper introduces a neural-dynamics-based active steering control (NDASC) scheme developed under artificial systems, computational experiments, and parallel execution (ACP) framework, aimed at enhancing the stability and reliability of autonomous vehicles in noisy environments. Based on the Taylor expansion theorem, noises can be represented in the form of polynomials for the desired accuracy, and therefore polynomial noises can be viewed as a more generalized representation of noises. Then, the proposed NDASC scheme includes a model predictive active steering control (MPASC) strategy solved by a polynomial noise resilience neural dynamics (PNRND) model. Computational experiments parallelly implemented upon the CarSim-Simulink platform substantiate the effectiveness and robustness of the proposed NDASC scheme, providing significant theoretical and practical insights for control strategies of autonomous vehicles under various noisy environments.

Keywords: Automated Vehicles
Abstract: In the area of vehicular ad hoc networks (VANETs), efficient data transmission stands as a cornerstone for ensuring dependable communication. This paper focuses on the key role of optimizing antenna configuration concerning data volumes to improve communication efficiency within VANETs. To attain efficient data transmission and improve user experience, we analyze and compare the influence of varied data volumes and antenna settings on communication quality stemming from bit error rate performance. Through experimental validation across diverse signal detection scenarios, disparities in performance across different antenna configurations are confirmed, thereby furnishing invaluable insights into the efficacy of dynamically selecting antenna configurations based on data volumes. In essence, this paper aims to demonstrate the intricacies of optimizing antenna settings to adaptly navigate fluctuations in transmitted data volumes, thereby increasing communication efficiency and ensuring robust communication in VANETs.

Keywords: Automated Vehicles
Abstract: Vehicle Re-identification (Re-ID) has been broadly studied in the last decade; however, the different camera view angles leading to confused discrimination in the feature subspace for the vehicles of various poses, is still challenging for the Vehicle Re-ID models in the real world. To promote the Vehicle Re-ID models, this paper proposes to synthesize a large number of vehicle images in the target pose, whose idea is to project the vehicles of diverse poses into the unified target pose so as to enhance feature discrimination. Considering that the paired data of the same vehicles in different traffic surveillance cameras might be not available in the real world, we propose the first Pair-flexible Pose Guided Image Synthesis method for Vehicle Re-ID, named as VehicleGAN in this paper, which works for both supervised and unsupervised settings without the knowledge of geometric 3D models. Because of the feature distribution difference between real and synthetic data, simply training a traditional metric learning based Re-ID model with data-level fusion (data augmentation) is not satisfactory, therefore we propose a new Joint Metric Learning (JML) via effective feature-level fusion from both real and synthetic data. Intensive experimental results on the public VeRi-776 and VehicleID datasets prove the accuracy and effectiveness of our proposed VehicleGAN and JML.

Keywords: Automated Vehicles
Abstract: This paper proposed a timescale graph-parallel (GP) computation method to solve the real-time optimization problem of nonlinear predictive energy-saving control, thus to realize the implementation of MPC on vehicle on-board controllers. The proposed scheme consists of two parts: forward prediction of the objective function and backpropagation of the partial differential function, both of which can be calculated in parallel. Thus, compared with traditional serial solution method for optimization problems, the timescale graph-parallel computation method can utilize the computing resources of the controller fully. In this paper, firstly, based on the characteristics of commercial vehicles, a mixed integral optimal control problem (MIOCP) was constructed. Then, a detailed timescale graph-parallel computation algorithm was derived for the MIOCP. Finally, GP and Pontryagin’s Minimum Principle (PMP) algorithms were applied on the predefined road for the simulation of the prediction of energy-saving control for commercial vehicles. The simulation results showed that compared with PMP, the maximum iteration number, average iteration number, single longest solution time, and single average solution time of the proposed GP decreased by 60%, 64.28%, 89.53%, and 93.56%, respectively. In addition, GP can also improve fuel efficiency by 1.55% without sacrificing much power performance.

Keywords: Automated Vehicles
Abstract: This paper introduces an innovative hybrid approach for estimating vehicle lateral velocity, merging mechanism-based methods with a Long Short-Term Memory (LSTM) neural network. Traditional estimation techniques, which are often susceptible to drift and inaccuracies due to parameter mismatches, fail to effectively adapt to varying driving conditions. Our proposed approach leverages the accuracy of mechanism-based estimates in specific scenarios to feed the LSTM network, creating a data-mechanism-driven solution. To overcome the inherent challenges of data-driven models, particularly concerning data quality and volume, our lateral velocity estimation model incorporates a feature extraction layer alongside a regression output layer. This architecture not only facilitates efficient parameter optimization within the feature extraction phase but also enables targeted retraining of the regression layer, significantly boosting transfer learning capabilities. We validate the robustness and the practicality of transfer learning across different vehicle classes in a simulation environment, showcasing its broad applicability and effectiveness.

Keywords: Automated Vehicles
Abstract: Detection of road obstacles is important for autonomous driving. However, road obstacles, like pedestrians, usually account for quite a small portion compared with other semantics, such as road layouts. This leads to the class-imbalance problem in real-world driving datasets and hinders environment perception for autonomous driving. In this paper, we propose an obstacle-sensitive network to improve the semantic Bird-Eye-View (BEV) map generation performance for minority classes. To this end, a context-depth attention module and a boundary-aware loss are introduced. We conduct ablation studies to verify the effectiveness of the proposed network. We also compare our network with other semantic BEV map generation methods. The results demonstrate that our network achieves better performance in terms of semantic BEV map generation, especially for minority classes.

Keywords: Automated Vehicles
Abstract: Traffic accidents are a significant factor leading to injuries and property losses, prompting extensive research in the field of traffic safety. However, previous studies, whether focused on static environment assessment, dynamic driving analysis, pre-accident prediction, or post-accident rule checks, have often been conducted independently. Our introduces V2X Environmental Perception Multi-modal Large Model AccidentGPT for accident analysis and prevention. AccidentGPT establishes a multi-modal information interaction framework based on multisensory perception. It adopts a holistic approach to address traffic safety issues, providing environmental perception for autonomous vehicles to avoid collisions and maintain control. In human-driven vehicles, it offers proactive safety warnings, blind spot alerts, and driving suggestions through human-machine dialogue. Additionally, it aids traffic police and management agencies in considering factors such as pedestrians, vehicles, roads, and the environment for intelligent real-time analysis of traffic safety. The system also conducts a thorough analysis of accident causes and post-accident liabilities, making it the first large-scale model to integrate comprehensive scene understanding into traffic safety research. Project page: https://accidentgpt.github.io

Keywords: Automated Vehicles
Abstract: Autonomous vehicles have stridden over the budding stage and are stepping into the phase of large-scale commercial deployment. Nonetheless, safety issues of autonomous driving remain to be fully solved. Sensor data provide the observations of the internal status and the driving environment of the autonomous vehicle, and sensor data anomaly detection is indispensable to ensure the safety since the occurrence of sensor data anomalies indicate potential safety risks. Tremendous works has contributed to the sensor data anomaly detection issue but most of them ignore the trustworthiness estimation of the anomaly detection results, leading to difficulties for decision-making in safety-critical systems. Therefore, this work proposes an uncertainty-aware sensor data anomaly detection method to enhance the trustworthiness of anomaly detection results. Specifically, this method includes a Bayesian LSTM prediction network that outputs both the predicted values and the distribution of the predicted values, an anomaly uncertainty quantification method, and an adaptive thresholding method to improve the anomaly detection performance. Anomaly detection is achieved by capturing the predicted values with high uncertainty. The efficacy and robustness of the proposed methodology have been substantiated through empirical field tests conducted with real-world autonomous driving vehicles. The evaluation yielded a recall of 0.893 and an F1-Score of 0.937, which underscores the superior anomaly detection capabilities of the approach within practical autonomous driving contexts.

Keywords: Automated Vehicles
Abstract: Vehicle trajectory prediction is crucial for ensuring the safety and reliability of autonomous driving systems. Due to the highly stochastic nature of road participants' behaviors, it is vital that prediction models accommodate a wide range of possible scenarios to mitigate safety risks. To address this challenge, we propose a novel trajectory prediction model called DiffusionTrajPred, an innovative trajectory prediction model based on the diffusion model. This model uniquely combines forward and reverse processes, manipulating noise levels in trajectory data to forecast future paths. Through the application of a mask-based reverse process, the model can make full use of historical trajectory information and predict trajectories that combine accuracy and multiple possibilities. The model utilizes a Transformer architecture for learning the noise, which enables the model to extract richer temporal information from trajectory data, resulting in improved semantic comprehension. Furthermore, we have effectively encoded high-definition (HD) semantic map information and vehicle interaction dynamics as crucial input features, improving the model's predictive power. This approach not only respects physical constraints but also elevates the accuracy of the predictions. Extensive experiments on the widely recognized open-source dataset 'Argoverse' reveal that our method outperformed the most existing state-of-the-art methods in terms of accuracy and multimodality, demonstrating the diffusion model's unique advantage in addressing the stochastic nature of road scenarios in autonomous driving.

Keywords: Automated Vehicles
Abstract: Recent research has demonstrated that the Vehicle-to-Everything (V2X) communication techniques can fundamentally improve the perception system for autonomous driving by collaborating between vehicle and infrastructure sensors. LiDAR is the commonly-used sensor for V2X autonomous driving due to its robustness in challenging scenarios. However, the LiDAR sensor is expensive, so the cost of equipping LiDAR sensors to a large number of infrastructures on the large-scale roadway network is extremely high, which has limited the wide deployment of the V2X cooperative perception system. How to discover an economic V2X cooperative perception system is never been well studied before. Inspired by the cost difference of the various point cloud densities of LiDAR, we propose the first Density-Sensitive Infrastructure LiDAR benchmark for economic V2X cooperative perception, named V2X-DSI, in this paper. Using the proposed V2X-DSI benchmark, we analyze the effect of cooperative perception performance under different beam infrastructure LiDAR. We specifically assess three state-of-the-art methods, OPV2V, V2X-ViT, and CoBEVT, using our V2X-DSI dataset. The results indicate that varying beam infrastructure LiDAR sensors play a crucial role in influencing cooperative perception performance.

Keywords: Automated Vehicles
Abstract: Online relative pose estimation within constrained time frame is a critical challenge for object-level multi-agent cooperative perception. Specifically, its objective is to determine the relative translation and rotation of cooperating agents such that the detected objects are aligned. Current methodologies adopt a non-probabilistic approach to data association and a singular association hypothesis is assumed, resulting in overconfident pose estimates and diminished accuracy in ambiguous environments. A probabilistic relative pose estimation approach is proposed to directly address this limitation by jointly considering all the potential association hypotheses and their respective likelihoods. We construct a comprehensive Bayesian estimation problem encompassing data association and pose inference. An iterative message-passing algorithm is employed on the proposed factor graph to derive near-optimal and real-time relative pose estimates.Numerical studies verify the real-time performance and effectiveness of the proposed method.

Keywords: Automated Vehicles
Abstract: Due to the limitations of network communication conditions for online calling GPT, the onboard deployment of Large Language Models for autonomous driving is in need. In this paper, we propose Drive as Veteran, a fine-tuned LLaMA-7B model with driving tasks. A training set consisting of instructions, scenario descriptions and human-annotated driving tasks is established. Through LoRA fine-tuning, the capability of generating correct driving tasks of our model is demonstrated through a numerical experiment and the comparison to GPT-3.5 is presented. We show that smaller-sized Large Language Models could be deployed onboard with fast generation speed and high accuracy, which could serve as a core component for decision-making in autonomous driving.

Keywords: Automated Vehicles
Abstract: Correctly understanding surrounding environments is a fundamental capability for autonomous driving. Semantic forecasting of bird-eye-view (BEV) maps can provide semantic perception information in advance, which is important for environment understanding. Currently, the research works on combining semantic forecasting and semantic BEV map generation is limited. Most existing work focuses on individual tasks only. In this work, we attempt to forecast semantic BEV maps in an end-to-end framework for future front-view (FV) images. To this end, we predict depth distributions and context features for FV input images and then forecast depth-context features for the future. The depth-context features are finally converted to the future semantic BEV maps. We conduct ablation studies and create baselines for evaluation and comparison. The results demonstrate that our network achieves superior performance.

Keywords: Automated Vehicles
Abstract: With the rapid development of autonomous driving technology, the safety of driving systems has increasingly become the focus of attention. However, although many existing autonomous driving decision-making algorithms, such as deep reinforcement learning, demonstrate excellent performance, their decision-making processes lack interpretability and are opaque to users. To address this problem, this paper constructs an interpretable driving model from the perspective of human cognition, which can not only imitate human driving behavior through cognitive reinforcement learning methods, but also show better performance in driving experiments. In addition, the paper also proposes an analysis method for abnormal driving behavior, which provides a new idea for discovering potential unsafe behaviors during driving and exploring the possible impact of this behavior pattern on driving tasks.

Keywords: Automated Vehicles
Abstract: In the digital era, data has become a pivotal asset, propelling the development of technologies such as autonomous driving. Despite this, data trading faces challenges of the absence of robust pricing methods and the lack of trustworthy trading mechanisms. To address these challenges, we introduce a traffic-oriented data trading platform named Data on The Move (DTM), integrating traffic simulation, data trading, and Artificial Intelligent (AI) agents. The DTM platform supports evident-based data value evaluation and AI-based trading mechanisms. Leveraging the common sense capabilities of Large Language Models (LLMs) to assess traffic state and data value, DTM can determine reasonable traffic data pricing through multi-round interaction and simulations. Moreover, DTM provides a pricing method validation by simulating traffic systems, multi-agent interactions, and the heterogeneity and irrational behaviors of individuals in the trading market. Within the DTM platform, entities such as connected vehicles and traffic light controllers could engage in information collecting, data pricing, trading, and decision-making. Simulation results demonstrate that our proposed AI agent-based pricing method facilitates data trading by giving out reasonable prices, which is evident by the traffic efficiency improvement. This underscores the effectiveness and practical value of DTM, offering new perspectives and tools for the evolution of data markets and smart city development. To the best of our knowledge, this is the first study employing LLMs in data pricing and a pioneering data trading practice in the field of intelligent vehicles and smart cities.

Keywords: Automated Vehicles
Abstract: This study constructs a three-degree-of-freedom (3-DOF) dynamics model for a Self-Propelled Modular Transporter (SPMT) with six axles. A novel particle swarm optimization with variable control factors and elitist learning (PSO-VFEL) is developed to quantify unknown terms and is compared with three existing optimizers, including an initial PSO, a traditional PSO, and a PSO with variable control factors (PSO-VF). Compared to the existing optimizers, the PSO-VFEL demonstrates higher flying flexibility and particle diversity, attributed to the generationally variable factors and additional guidance from the elitist learning strategy. Simulation data from the Carla software is employed for model calibration. Results show that the developed PSO-VFEL is superior for calibrating the dynamics model, achieving fitness improvements of 31.852%, 20.231%, and 5.154% compared to the initial PSO, the traditional PSO, and the PSO-VF, respectively. This research provides a dynamics model and a complementary calibrator for engineering applications.

Keywords: Automated Vehicles
Abstract: Cooperative perception algorithms based on fusing sensing data across multiple connected automated vehicles (CAVs) have shown promising performance to enhance the existing individual perception algorithm in terms of object detection tracking. However, existing cooperative perception algorithms are only developed offline given the constraints from data sharing and computational resources and none of them have been verified in real-time conditions. In this work, we propose a real-time cooperative perception framework called CooperFuse, which achieves cooperative perception in a late fusion scheme. Specifically, we handle online time and spatial synchronization by introducing a time calibration module and LiDAR-inertia-based localization component. Additionally, based on object detection and tracking results from individual vehicle perception, we devise a cooperative perception algorithm in a late fusion framework that considers object detection confidence score, kinematics, and dynamics consistency as well as scale consistency of detected objects. The algorithm computes the kinematic and dynamic consistency of the objects by solving for the energy consumption of inter-frame trajectories, and determines scale consistency by calculating inter-frame scale changes, enabling feature-based bounding box fusion. The experimental results demonstrate the real-time performance of the proposed algorithm and reveal its effective improvements in feature fusion and object detection accuracy when dealing with heterogeneous detection models across different cooperative intelligent agents.

Keywords: Automated Vehicles
Abstract: In this paper, we propose a cloud control method with communication delay prediction for intelligent connected vehicles (ICVs), which not only constructs a prediction model using real-world 5G communication delay data, but also evaluate the effectiveness of the cloud control method considering delay in typical application scenarios. Firstly, for the application data interaction of 5G vehicle-to-network-to-vehicle (V2N2V) full-link, we collect a large amount of communication delay data through vehicle test. Then, a novel data-driven delay prediction method based on the Long Short-Term Memory (LSTM) network is introduced. Finally, a cloud control method considering communication delay is constructed at unsignalized intersection. The test results show that our method can not only achieve high delay prediction accuracy, but also significantly reduce vehicle velocity fluctuations and avoid collisions.

Keywords: Automated Vehicles
Abstract: Vehicle detection in Unmanned Aerial Vehicle (UAV) captured images has wide applications in aerial photography and remote sensing. There are many public benchmark datasets proposed for the vehicle detection and tracking in UAV images. Recent studies show that adding an adversarial patch on objects can fool the well-trained deep neural networks based object detectors, posing security concerns to the downstream tasks. However, the current public UAV datasets might ignore the diverse altitudes, vehicle attributes, fine-grained instance-level annotation in mostly side view with blurred vehicle roof, so none of them is good to study the adversarial patch based vehicle detection attack problem. In this paper, we propose a new dataset named EVD4UAV as an altitude-sensitive benchmark to evade vehicle detection in UAV with 6,284 images and 90,886 fine-grained annotated vehicles. The EVD4UAV dataset has diverse altitudes (50m, 70m, 90m), vehicle attributes (color, type), fine-grained annotation (horizontal and rotated bounding boxes, instance-level mask) in top view with clear vehicle roof. One white-box and two black-box patch based attack methods are implemented to attack three classic deep neural networks based object detectors on EVD4UAV. The experimental results show that these representative attack methods could not achieve the robust altitude-insensitive attack performance.

Keywords: Automated Vehicles
Abstract: Cooperative driving automation attracts great attention for its potential to improve traffic safety. Knowing each vehicle's accurate position serves as the cornerstone for the information fusion necessary in cooperative driving tasks. However, inherent errors within a vehicle's self-localization system often necessitate correction to facilitate cooperative perception and downstream tasks. Leveraging intermediate features shared among other Connected Automated Vehicles (CAVs), we propose an end-to-end learning localization framework aimed at estimating the relative pose error between the ego vehicle and the CAV. We investigate factors that may influence learning performance and validate the algorithm's performance using a simulation dataset. The proposed method is compared with the traditional point cloud matching-based relative localization method. Remarkably, our framework effectively corrects relative pose errors even when the vehicle exhibits significant initial localization inaccuracies, and it can be integrated into the cooperative perception system.

Keywords: Automated Vehicles
Abstract: The advent of Large Language Models (LLM) provides new insights to validate Automated Driving Systems (ADS). In the herein-introduced work, a novel approach to extracting scenarios from naturalistic driving datasets is presented. A framework called Chat2Scenario is proposed leveraging the advanced Natural Language Processing (NLP) capabilities of LLM to understand and identify different driving scenarios. By inputting descriptive texts of driving conditions and specifying the criticality metric thresholds, the framework efficiently searches for desired scenarios and converts them into ASAM OpenSCENARIO and IPG CarMaker text files. This methodology streamlines the scenario extraction process and enhances efficiency. Simulations are executed to validate the efficiency of the approach. The framework is presented based on a user-friendly web app and is accessible via the following link: https://github.com/ftgTUGraz/Chat2Scenario.

Keywords: Automated Vehicles
Abstract: Curve Speed Warning (CSW) systems assist drivers in adjusting speeds before entering a curve to improve road safety. As an essential part of CSW, the safe speed model is key in determining the speed trajectory. Current safe speed models are mostly based on the theoretical line shape of the road, which leads to the neglect of the driving differences, and it is likely to result in unreasonable speed guidance. This paper proposes a more comprehensive method to provide a safe speed trajectory in advance and enhance safety for trucks with heavy loads when approaching curve-slope sections. First, a classification model applying a random forest algorithm is developed to output the critical safe speed in a specific scenario. Second, a variable speed limit algorithm for a given path is framed, minimizing fuel and travel time consumption, and then embedded with a variable speed limit determination process. Simulation experiments are implemented based on real-world paths to verify the proposed structure. The findings indicate that our model is capable of generating speed trajectories adaptively. Additionally, experiments underscore the significant influence that the weight of the load and its center of gravity (CG) exert on the stability assessment of trucks, as we conclude that the optimal loading strategy for trucks is to reach a full load and avoid the load's lateral offset.

Keywords: Automated Vehicles
Abstract: Motion forecasting plays a pivotal role in autonomous driving systems, enabling vehicles to execute collision warnings and rational local-path planning based on predictions of the surrounding vehicles. However, prevalent methods often assume complete observed trajectories, neglecting the potential impact of missing values induced by object occlusion, scope limitation, and sensor failures. Such oversights inevitably compromise the accuracy of trajectory predictions. To tackle this challenge, we propose an end-to-end framework, termed Multiscale Transformer (MSTF), meticulously crafted for incomplete trajectory prediction. MSTF integrates a Multiscale Attention Head (MAH) and an Information Increment-based Pattern Adaptive (IIPA) module. Specifically, the MAH component concurrently captures multiscale motion representation of trajectory sequence from various temporal granularities, utilizing a multi-head attention mechanism. This approach facilitates the modeling of global dependencies in motion across different scales, thereby mitigating the adverse effects of missing values. Additionally, the IIPA module adaptively extracts continuity representation of motion across time steps by analyzing missing patterns in the data. The continuity representation delineates motion trend at a higher level, guiding MSTF to generate predictions consistent with motion continuity. We evaluate our proposed MSTF model using two large-scale real-world datasets. Experimental results demonstrate that MSTF surpasses state-of-the-art (SOTA) models in the task of incomplete trajectory prediction, showcasing its efficacy in addressing the challenges posed by missing values in motion forecasting for autonomous driving systems.

Keywords: Automated Vehicles
Abstract: This paper explores the intersection of recent AI advancements and Intelligent Transportation Systems (ITS), specifically focusing on enhancing the capabilities of Connected and Automated Vehicles (CAVs) in dynamic traffic scenarios. While combinations of vehicular sensors and AI offer promising prospects for advanced environmental perception, challenges still persist in accurately identifying dangers during the transition to automated traffic. The ESRIUM project, funded by the EU Horizon 2020 Programme, aims to address these challenges by developing digital maps representing road deterioration and employing Vehicle-to-Everything (V2X) communication to generate infrastructure-assisted routing recommendations for CAVs. While the solutions for sending standardized safety messages and controlling enabled CAVs were demonstrated in the ESRIUM project, the solution for the automatic generation of Cooperative Intelligent Transport Systems (C-ITS) safety messages was not studied. In this paper, we propose a pipeline named ``ConnectGPT'', which connects Large Language Models (LLMs) with CAVs, utilizing GPT-4, to observe traffic conditions, identify conditions that can endanger the flow of traffic, and automate the generation of the corresponding standardized C-ITS messages, such as Decentralized Environmental Notification Message (DENM) about the actual safety problem. Practical experiments with ongoing development show potential for real-world applications, which can significantly improve traffic management efficiency and enhance the security of all traffic participants, marking a crucial advancement in the integration of AI tools in ITS.

Keywords: Automated Vehicles
Abstract: Accurately forecasting the future paths of numerous agents is vital for the efficacy of autonomous systems. In crowded scenarios such as sidewalks, subways and airports, pedestrians instinctively modify their motion pattern in response to the environmental context and social consensus like preserving personal space and circumventing physical contact. Thus, the task to predict future pedestrian trajectory presents considerable challenges owing to the complex interaction among agents and the inherent uncertainty in predicting each agent's subsequent actions. Inspired by the recent success of Graph Neural Networks (GNN), a model named DATraj is introduced for predicting pedestrian trajectory. DATraj first uses a temporal encoder composed of attention mechanism to capture the spatial-temporal dynamics of pedestrians. The encoder can learn the motion pattern and subtle movement of pedestrian in the crowded scenario. Graph Attention Networks (GAT) is used in many models to catch social interaction between individuals. However common GATs compute a static attention: the ranking of the attention scores is unconditioned on the query node. DATraj implement the global interaction parts using the improved dynamic attention which every query uniquely prioritizes the attention coefficients correlated with the keys, this provides a much better robustness to noise. Experiments show that our trajectory prediction model achieves better performance on several public datasets.

Keywords: Automated Vehicles
Abstract: Achieving top-notch performance in Intelligent Transportation detection is a critical research area. However, many challenges still need to be addressed when it comes to detecting in a cross-domain scenario. In this paper, we propose a Self-Aware Adaptive Alignment (SA3), by leveraging an efficient alignment mechanism and recognition strategy. Our proposed method employs a specified attention-based alignment module trained on source and target domain datasets to guide the image-level features alignment process, enabling the localglobal adaptive alignment between the source domain and target domain. Features from both domains, whose channel importance is re-weighted, are fed into the region proposal network, which facilitates the acquisition of salient region features. Also, we introduce an instance-to-image level alignment module specific to the target domain to adaptively mitigate the domain gap. To evaluate the proposed method, extensive experiments have been conducted on popular cross-domain object detection benchmarks. Experimental results show that SA3 achieves superior results to the previous state-of-the-art methods.

Keywords: Automated Vehicles
Abstract: In recent years, research on Vehicle-to-Everything (V2X) cooperative perception algorithms mainly focuses on the fusion of intermediate features from LiDAR point clouds. Since the emergence of excellent single-vehicle visual perception models like BEVFormer, collaborative perception schemes based on camera and late-fusion have become feasible approaches. This paper proposes a V2X-collaborative 3D object detection structure in Bird’s Eye View (BEV) space, based on global non-maximum suppression and late-fusion (V2X-BGN), and conducts experiments on the V2X-Set dataset. Focusing on complex road conditions with extreme occlusion, the paper compares the camera-based algorithm with the LiDAR-based algorithm, validating the effectiveness of pure visual solutions in the collaborative 3D object detection task. Additionally, this paper highlights the complementary potential of camera-based and LiDAR-based approaches and the importance of object-to-ego vehicle distance in the collaborative 3D object detection task.

Keywords: Automated Vehicles
Abstract: Learning and understanding car-following (CF) behaviors are crucial for microscopic traffic simulation. Traditional CF models, though simple, often lack generalization capabilities, while many data-driven methods, despite their robustness, operate as “black boxes” with limited interpretability. To bridge this gap, this work introduces a Bayesian Matrix Normal Mixture Regression (MNMR) model that simultaneously captures feature correlations and temporal dynamics inherent in CF behaviors. This approach is distinguished by its separate learning of row and column covariance matrices within the model framework, offering an insightful perspective into the human driver decision-making processes. Through extensive experiments, we assess the model’s performance across various historical steps of inputs, predictive steps of outputs, and model complexities. The results consistently demonstrate our model’s adeptness in effectively capturing the intricate correlations and temporal dynamics present during CF. A focused case study further illustrates the model’s outperforming interpretability of identifying distinct operational conditions through the learned mean and covariance matrices. This not only underlines our model’s effectiveness in understanding complex human driving behaviors in CF scenarios but also highlights its potential as a tool for enhancing the interpretability of CF behaviors in traffic simulations and autonomous driving systems.

Keywords: Automated Vehicles
Abstract: In natural disasters, the emergency rescue system is of utmost importance for the smart city development, but the search and rescue (SAR) leveraging unmanned aerial vehicles (UAVs) is under-explored. This study has established an Internet of UAVs architecture motivated by search and rescue activities, focusing on the path planning problem of UAV in three-dimensional urban disaster scenes. A mathematical model for UAV-based SAR missions has been built, aiming to use UAV platforms equipped with life-detection radars to search for survivors inside buildings. Our aim is to search for the most survivors in the shortest amount of time while ensuring coverage of the entire search area. Based on the mathematical model for SAR activities, the improved Multi-Verse Optimizer (IMVO) is used to solve the path of UAV. Finally, simulations were conducted in a photorealistic urban scenario, demonstrating the paths generated by the proposed method. The results indicate the potential of the generated path in terms of search time and the number of survivors discovered.

Keywords: Automated Vehicles
Abstract: Testing and evaluating the safety performance of autonomous vehicles (AVs) is essential before the large-scale deployment. Practically, the number of testing scenarios permissible for a specific AV is severely limited by tight constraints on testing budgets and time. With the restrictions imposed by strictly restricted numbers of tests, existing testing methods often lead to significant uncertainty or difficulty to quantifying evaluation results. In this paper, we formulate this problem for the first time the “few-shot testing” (FST) problem and propose a systematic framework to address this challenge. To alleviate the considerable uncertainty inherent in a small testing scenario set, we frame the FST problem as an optimization problem and search for the testing scenario set based on neighborhood coverage and similarity. Specifically, under the guidance of better generalization ability of the testing scenario set on AVs, we dynamically adjust this set and the contribution of each testing scenario to the evaluation result based on coverage, leveraging the prior information of surrogate models (SMs). With certain hypotheses on SMs, a theoretical upper bound of evaluation error is established to verify the sufficiency of evaluation accuracy within the given limited number of tests. The experiment results on cut-in scenarios demonstrate a notable reduction in evaluation error and variance of our method compared to conventional testing methods, especially for situations with a strict limit on the number of scenarios.

Keywords: Automated Vehicles
Abstract: Traffic oscillations pose safety and efficiency challenges in mixed scenarios involving connected and automated vehicles (CAVs) and human-driven vehicles (HDVs). Existing control strategies fail to handle the unpredictability of HDV behaviors, resulting in disruptive "stop-and-go" traffic patterns. This study proposes a novel algorithm that uses Deep Reinforcement Learning (DRL) integrated into a distinctive "CAV-AHDV-CAV" structure for car-following events. The consecutive HDVs are treated as an aggregated unit called Aggregated HDVs (AHDVs) to eliminate stochasticity and leverage collective traffic features as inputs, addressing the driver heterogeneity issue. Our training and testing were conducted using a dataset of 9,200 car-following events extracted from the HighD dataset. In these events, the lead vehicle serves as our CAV in front, while the following vehicle represents the AHDV. We simulated our controlled vehicle to follow the AHDV, aiming to achieve the vehicle equilibrium state with respect to both the AHDV and the CAV in front. The results demonstrate a reduction in the impact of HDVs and an enhancement of equilibrium states compared to baseline models. Specifically, we achieved a speed mean square error (MSE) of 3.151 and spacing MSE values of 50.484 (with respect to the AHDV) and 47.855 (with respect to the CAV). These findings offer robust and adaptable control strategies for efficient and safe mixed traffic dominated by CAVs.

Keywords: Automated Vehicles
Abstract: With the rapid advancement of deep learning technology, feature extraction backbones that are effectively trained have found increasing use in various traffic perception tasks, such as vehicle recognition and roadway user detection and classification. However, given the naturally imbalanced distribution of objects in the real world, deep learning networks can inadvertently act as bias amplifiers, leading to unfair detection and classification outcomes. Addressing and quantifying this bias in traffic applications has thus become a pressing challenge. In response, this research introduces the first comprehensive traffic imbalance object recognition dataset tailored for autonomous vehicles, called the Autonomous-vehicle Long-tail Image Dataset (ALIDA). This dataset reflects real-world sample distribution and includes four categories—motorized users, non-motorized users, roadway facilities, and traffic signs—spanning 87 classes and totaling 37,558 images. Our experimental results confirm that these backbones may struggle to accurately recognize less common objects with limited training data, such as children and wheelchair users. To mitigate such biases and improve traffic perception equality, we introduce a DEbiased Traffic Object Recognition (DETOR) scheme. This scheme leverages both few-shot and representation learning techniques. Employing DETOR, the residual neural network achieved a 290% increase in accuracy for recognizing minority classes, such as children, motorcyclists, deer, and bears. This not only enhances the effectiveness but also significantly improves the fairness and scalability of traffic perception using deep neural networks.

Keywords: Automated Vehicles
Abstract: In this paper, we propose the innovative use of Intelligent Vehicle Carriers (IVCs) as a key solution to address the energy constraints of small-scale unmanned Intelligent Vehicles (IVs). IVCs function as both transporters and charging stations, significantly boosting the operational range and efficiency of IVs. Our research delves into the IV-IVC collaborative framework, highlighting the existing challenges, exploring potential solutions, and examining a range of applications. This study offers a visionary approach to revolutionizing intelligent transportation systems by leveraging the synergistic relationship between IVs and IVCs.

Keywords: Automated Vehicles
Abstract: Vehicle recognition and classification are critical for numerous traffic applications such as traffic signal control, traffic flow modeling, tolling, and logistics optimization. Common sensing systems often rely on in-pavement loops or surveillance video cameras, both of which have inherent limitations. This study introduces the Compact LiDAR Empowered Vehicle Enhancing-minority Recognition (CLEVER) system, a real-time, cost-effective vehicle detection and classification framework powered by edge Artificial Intelligence (AI). Utilizing a customized minority-enhancing vehicle classification deep neural network, the CLEVER system surpasses existing LiDAR-based vehicle classification methods, achieving up to a 15.98% true-positive rate in classifying ten types of vehicles. By integrating the hardware, pre-processing algorithms, and classification neural network into an edge computing node, the CLEVER system reduces the costs associated with LiDAR systems by 90% and operates effectively in a plug-and-play mode with a negligible sub-second inference time (212ms to 459ms). The CLEVER system provides an affordable end-to-end solution that enhances the accuracy and reliability of vehicle classification data, potentially leading to more efficient and flexible Intelligent Transportation Systems (ITS).

Keywords: Automated Vehicles
Abstract: Model-based and learning-based methods are two main approaches modeling car-following behaviors. To combine advantages from both types of models, this study introduces a novel approach, IDM-Follower, which generates a sequence of the following vehicle's trajectory using a recurrent autoencoder informed by a physical car-following model, the Intelligent Driving Model (IDM). We design an innovative neural network (NN) structure with two independent encoders and an attention-based decoder to predict the trajectory sequence. The loss function accounts for discrepancies from both the physical car-following model and the NN predictions. Numerical experiments are conducted using simulated and real world (i.e., NGSIM) datasets under different data noise levels with varying weights between the learning loss and the model loss. Testing results show the proposed approach outperforms both model-based and learning-based baselines under real and high noise levels. The optimal integrating weight between the model and learning component is significantly influenced by data quality, which affects both prediction accuracy and safety metrics.

Keywords: Automated Vehicles
Abstract: Tutorial: IEEE ITS Standards Development and Update

16:30-16:40 Dr. Meng Lu (IEEE ITSS/SC): ITSS/SC and Tutorial Introduction

16:40-17:05 Soo Kim (IEEE SA): IEEE SA and IEEE Standards

17:05-17:30 Dr. Ignacio Alvarez (Intel Labs, USA) & Dr. Javier Ibanez-Guzman (Renault S.A., France): CAD Deployment and Standards

17:30-18:20 Discussions / Q&A

18:20-18:30 Dr. Javier Ibanez-Guzman (IEEE ITSS/SC): Wrap-up

Note: The Tutorial at 16:30-18:30 will be hybrid, and one can join online.

JOIN WEBEX MEETING https://ieee.webex.com/ieee/j.php?MTID=mcbc296e26c474456b8a 8239f60b64205 Meeting number (access code): 2634 853 1835

Meeting password: pBVSE9XYe32