Park, Ji Hwan (The University of Texas at Austin), Zhang, Yanze (University of North Carolina at Charlotte), Luo, Wenhao (University of North Carolina at Charlotte), Wang, Junmin (The University of Texas at Austin)

Individualizable Risk Assessment Map for Planning Automated Vehicle Behaviors Respecting Perceived Safety

Scheduled for presentation during the Invited Session "Safety for Intelligent and Connected Vehicles" (ThBT3), Thursday, September 26, 2024, 15:10−15:30, Salon 6

2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), September 24- 27, 2024, Edmonton, Canada

This information is tentative and subject to change. Compiled on October 3, 2024

Abstract

With the continual introduction of various vehicle automation functions and algorithms, the heterogeneity of human, human-machine, and machine autonomy associated with the vehicles on the roads elevates as well. Ensuring vehicles of heterogeneous autonomy to safely and harmoniously share the roads is important for the interests of driving safety, traffic throughput, and human acceptance of autonomy. Particularly due to the wide range of variations in human perceived safety and risk in driving, empowering the vehicle intelligence to accurately understand the perceived risk by other surrounding vehicles is a critical step towards harmonious road sharing. In this paper, we present an individualizable risk assessment model (IRAM) that can empower vehicles to compute the risk maps that individual drivers perceive from the driving behaviors of other surrounding vehicles on a shared road. The IRAM offers a realistic risk assessment map around a vehicle by integrating the relative motions between pertinent vehicles and drivers’ perceived safety preferences. A naturalistic human driving dataset, NGSIM, was utilized to evaluate the risk map generation by the proposed IRAM. The IRAM can be used for planning vehicle behaviors, such as its path and motion, that respect the different perceived safety of its surrounding vehicles’ human/machine drivers.