ITSC 2025 Paper Abstract

Paper FR-LA-T44.2

Liu, Yicai (Tsinghua University), Li, Yushu (College of Energy and Power Engineering, Nanjing University of ), Pan, Lei (Yanshan University), Wang, Xiangyu (Tsinghua University), Li, Liang (Tsinghua University), Fan, Yihong (https://its.papercept.net/conferences/scripts/pinwizard.pl), Zhang, Guowang (State Key Laboratory of Automotive Safety and Energy, Tsinghua U)

Smooth Human-Machine Takeover Via Neurodynamic Torque-Angle Switching Control in Steer-By-Wire Systems

Scheduled for presentation during the Regular Session "S44c-Human Factors and Human Machine Interaction in Automated Driving" (FR-LA-T44), Friday, November 21, 2025, 16:20−16:40, Currumbin

2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), November 18-21, 2025, Gold Coast, Australia

This information is tentative and subject to change. Compiled on October 18, 2025

Keywords Driver Behavior Monitoring and Feedback Systems for Semi-autonomous Vehicles, User-Centric HMI Design for Autonomous Vehicle Control Systems, Human-Machine Interaction Systems for Enhanced Driver Assistance and Safety

Abstract

Human-machine takeover is a critical function in conditionally automated driving, requiring smooth transitions between manual and automated control. However, challenges remain in achieving consistent road sense, switchable control strategies, and smooth takeover processes. To this end, this paper proposes a novel smooth takeover framework for steer-by-wire (SbW) systems, featuring a dual-objective controller based on model predictive control (MPC) that supports both torque and angle targets. Initially, the redundant SbW system configuration is introduced and modeled as the foundation of the controller. Subsequently, the dual-target controller is developed with disturbance compensation and adaptive weight adjustment to guarantee precision and smoothness in the tracking processes. The neurodynamic optimization algorithm is then proposed, whose lightweight computation can ensure real-time performance in embedded control units. Real-vehicle experiments validate the overall framework in terms of tracking accuracy, transition smoothness, and real-time feasibility, whose performances have been recognized and are being actively deployed in collaboration with industrial partners.