TY - GEN
T1 - Inverse Reinforcement Learning for Personalized and Safe Trajectory Planning in Autonomous Driving
AU - Wu, Chengye
AU - Zhang, Nianhua
AU - Wei, Henglai
AU - Li, Yan
AU - Zhang, Hui
N1 - Publisher Copyright:
© 2025 IEEE.
PY - 2025
Y1 - 2025
N2 - To enable human-like autonomous driving, we propose a framework that integrates personalization with safety. It first establishes safe trajectory boundaries using a multi-factor coupled field model. Then, it quantifies driving style by learning reward weights from driver decisions via Maximum Entropy Inverse Reinforcement Learning (MaxEnt IRL). Finally, these learned parameters guide polynomial trajectory planning within the safety constraints. Experiments validate the method's effectiveness in generating safe, stable, and personalized path planning.
AB - To enable human-like autonomous driving, we propose a framework that integrates personalization with safety. It first establishes safe trajectory boundaries using a multi-factor coupled field model. Then, it quantifies driving style by learning reward weights from driver decisions via Maximum Entropy Inverse Reinforcement Learning (MaxEnt IRL). Finally, these learned parameters guide polynomial trajectory planning within the safety constraints. Experiments validate the method's effectiveness in generating safe, stable, and personalized path planning.
KW - Maximum Entropy Inverse Reinforcement Learning (MaxEnt IRL)
KW - Multi-Factor Coupled Field
KW - Personalized Driving
KW - Trajectory Planning
UR - https://www.scopus.com/pages/publications/105041163317
U2 - 10.1109/CAC67268.2025.11487865
DO - 10.1109/CAC67268.2025.11487865
M3 - 会议稿件
AN - SCOPUS:105041163317
T3 - Proceedings - 2025 China Automation Congress, CAC 2025
SP - 7519
EP - 7523
BT - Proceedings - 2025 China Automation Congress, CAC 2025
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2025 China Automation Congress, CAC 2025
Y2 - 26 September 2025 through 28 September 2025
ER -