Developing reliable autonomous driving algorithms poses challenges in testing, particularly when it comes to safety-critical traffic scenarios involving pedestrians. An open question is how to simulate rare events, not necessarily found in autonomous driving datasets or scripted simulations, but which can occur in testing, and, in the end may lead to severe pedestrian related accidents. This paper presents a method for designing a suicidal pedestrian agent within the CARLA simulator, enabling the automatic generation of traffic scenarios for testing safety of autonomous vehicles (AVs) in dangerous situations with pedestrians. The pedestrian is modeled as a reinforcement learning (RL) agent with two custom reward functions that allow the agent to either arbitrarily or with high velocity to collide with the AV. Instead of significantly constraining the initial locations and the pedestrian behavior, we allow the pedestrian and autonomous car to be placed anywhere in the environment and the pedestrian to roam freely to generate diverse scenarios. To assess the performance of the suicidal pedestrian and the target vehicle during testing, we propose three collision-oriented evaluation metrics. Experimental results involving two state-of-the-art autonomous driving algorithms trained end-to-end with imitation learning from sensor data demonstrate the effectiveness of the suicidal pedestrian in identifying decision errors made by autonomous vehicles controlled by the algorithms.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Suicidal Pedestrian: Generation of Safety-Critical Scenarios for Autonomous Vehicles


    Beteiligte:
    Yang, Yuhang (Autor:in) / Kujanpaa, Kalle (Autor:in) / Babadi, Amin (Autor:in) / Pajarinen, Joni (Autor:in) / Ilin, Alexander (Autor:in)

    Erscheinungsdatum :

    2023


    Anmerkungen:

    6 pages; 5 figures; 2 tables


    Medientyp :

    Preprint


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Suicidal Pedestrian: Generation of Safety-Critical Scenarios for Autonomous Vehicles

    Yang, Yuhang / Kujanpaa, Kalle / Babadi, I Amin et al. | IEEE | 2023


    Adversarial Generation of Safety-Critical Lane-Change Scenarios for Autonomous Vehicles

    He, Zimin / Zhang, Jiawei / Yao, Danya et al. | IEEE | 2023


    Safety Effectiveness of Autonomous Vehicles and Connected Autonomous Vehicles in Reducing Pedestrian Crashes

    Susilawati, Susilawati / Wong, Wei Jie / Pang, Zhao Jian | Transportation Research Record | 2022



    Optimized deep learning for pedestrian safety in autonomous vehicles

    Farhat, Wajdi / Rhaiem, Olfa Ben / Faiedh, Hassene et al. | Elsevier | 2025

    Freier Zugriff