As autonomous driving technology matures, the safety and robustness of its key components, including trajectory prediction is vital. Although real-world datasets such as Waymo Open Motion provide recorded real scenarios, the majority of the scenes appear benign, often lacking diverse safety-critical situations that are essential for developing robust models against nuanced risks. However, generating safety-critical data using simulation faces severe simulation to real gap. Using real-world environments is even less desirable due to safety risks. In this context, we propose an approach to utilize existing real-world datasets by identifying safetyrelevant scenarios naively overlooked, e.g., near misses and proactive maneuvers. Our approach expands the spectrum of safety-relevance, allowing us to study trajectory prediction models under a safety-informed, distribution shift setting. We contribute a versatile scenario characterization method, a novel scoring scheme for reevaluating a scene using counterfactual scenarios to find hidden risky scenarios, and an evaluation of trajectory prediction models in this setting. We further contribute a remediation strategy, achieving a 10% average reduction in predicted trajectories’ collision rates. To facilitate future research, we release our code for this overall SafeShift framework to the public: github.com/cmubig/SafeShift


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    SafeShift: Safety-Informed Distribution Shifts for Robust Trajectory Prediction in Autonomous Driving


    Beteiligte:
    Stoler, Benjamin (Autor:in) / Navarro, Ingrid (Autor:in) / Jana, Meghdeep (Autor:in) / Hwang, Soonmin (Autor:in) / Francis, Jonathan (Autor:in) / Oh, Jean (Autor:in)


    Erscheinungsdatum :

    02.06.2024


    Format / Umfang :

    4437045 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Graph-based Planning-informed Trajectory Prediction for Autonomous Driving

    Dong, Qing / Jiang, Titong / Xu, Tao et al. | IEEE | 2022


    Advancing Autonomous Driving Safety Through LLM Enhanced Trajectory Prediction

    Cheng, Qian / Jiao, Xinyu / Yang, Mengmeng et al. | Springer Verlag | 2024

    Freier Zugriff

    Intention-Driven Trajectory Prediction for Autonomous Driving

    Fan, Shiwei / Li, Xiangxu / Li, Fei | IEEE | 2021


    INTENTION-DRIVEN TRAJECTORY PREDICTION FOR AUTONOMOUS DRIVING

    Fan, Shiwei / Li, Xiangxu / Li, Fei | British Library Conference Proceedings | 2021


    Trajectory-Based Failure Prediction for Autonomous Driving

    Kuhn, Christopher B. / Hofbauer, Markus / Petrovic, Goran et al. | British Library Conference Proceedings | 2021