Multi-task intersection navigation including the unprotected turning left, turning right, and going straight in dense traffic is still a challenging task for autonomous driving. For the human driver, the negotiation skill with other interactive vehicles is the key to guarantee safety and efficiency. However, it is hard to balance the safety and efficiency of the autonomous vehicle for multi-task intersection navigation. In this paper, we formulate a multi-task safe reinforcement learning with social attention to improve the safety and efficiency when interacting with other traffic participants. Specifically, the social attention module is used to focus on the states of negotiation vehicles. In addition, a safety layer is added to the multi-task reinforcement learning framework to guarantee safe negotiation. We compare the experiments in the simulator SUMO with abundant traffic flows and CARLA with high-fidelity vehicle models, which both show that the proposed algorithm can improve safety with consistent traffic efficiency for multi-task intersection navigation.


    Access

    Download


    Export, share and cite



    Title :

    Multi-task Safe Reinforcement Learning for Navigating Intersections in Dense Traffic


    Contributors:
    Liu, Yuqi (author) / Zhang, Qichao (author) / Zhao, Dongbin (author)

    Publication date :

    2022


    Type of media :

    Preprint


    Type of material :

    Electronic Resource


    Language :

    English




    A Multi-Task Reinforcement Learning Approach for Navigating Unsignalized Intersections

    Kai, Shixiong / Wang, Bin / Chen, Dong et al. | IEEE | 2020


    A MULTI-TASK REINFORCEMENT LEARNING APPROACH FOR NAVIGATING UNSIGNALIZED INTERSECTIONS

    Kai, Shixiong / Wang, Bin / Chen, Dong et al. | British Library Conference Proceedings | 2020


    Heterogeneous traffic intersections control design based on reinforcement learning

    Jiajing Shen / Jiaxing Hu / Qinpei Zhao et al. | DOAJ | 2024

    Free access

    Heterogeneous traffic intersections control design based on reinforcement learning

    Shen, Jiajing / Hu, Jiaxing / Zhao, Qinpei et al. | Wiley | 2024

    Free access

    Autonomously navigating across intersections

    BAALKE URIAH / BRAUN STAV / JIN SONIA | European Patent Office | 2022

    Free access