An open problem for autonomous driving is how to validate the safety of an autonomous vehicle in simulation. Automated testing procedures can find failures of an autonomous system but these failures may be difficult to interpret due to their high dimensionality and may be so unlikely as to not be important. This work describes an approach for finding interpretable failures of an autonomous system. The failures are described by signal temporal logic expressions that can be understood by a human, and are optimized to produce failures that have high likelihood. Our methodology is demonstrated for the safety validation of an autonomous vehicle in the context of an unprotected left turn and a crosswalk with a pedestrian. Compared to a baseline importance sampling approach, our methodology finds more failures with higher likelihood while retaining interpretability.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Interpretable Safety Validation for Autonomous Vehicles


    Contributors:


    Publication date :

    2020-09-20


    Size :

    696526 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Interpretable Multi-Task Prediction Neural Network for Autonomous Vehicles

    Wang, Qi / Hu, Hongyu / Yang, Biao et al. | IEEE | 2025


    INTERPRETABLE KALMAN FILTER COMPRISING NEURAL NETWORK COMPONENT(S) FOR AUTONOMOUS VEHICLES

    CARTER JOHN BRYAN / PAPI FRANCESCO / SONG QIAN et al. | European Patent Office | 2024

    Free access

    SOFTWARE VALIDATION FOR AUTONOMOUS VEHICLES

    MORLEY FRANKLIN / BAROR OMER / FAIRFIELD NATHANIEL et al. | European Patent Office | 2023

    Free access

    SOFTWARE VALIDATION FOR AUTONOMOUS VEHICLES

    MORLEY FRANKLIN / BAROR OMER / FAIRFIELD NATHANIEL et al. | European Patent Office | 2022

    Free access

    Software validation for autonomous vehicles

    MORLEY FRANKLIN / BAROR OMER / FAIRFIELD NATHANIEL et al. | European Patent Office | 2022

    Free access