Scenario-based testing is a common way to test automated and autonomous driving systems. Experts derive scenario types such as “left turn at intersection” and refine them into parameterized scenarios. The parameter domains span a huge space of possible test cases. Not all of these intuitively are “good” test cases in that they are of the correct form and challenge the driving system under test. For the selection of “good” test cases, the literature has proposed search-based techniques that are guided by optimization goals in the form of fitness functions. Existing works focus on technical aspects or create such fitness functions ad-hoc, leaving the methodological aspect of deriving suitable fitness functions unconsidered. There are suggestions regarding methodological guidance in the form of fitness function templates, but these ideas have so far been exclusively instantiated for straight sections on a highway. The methodological challenge of how to derive fitness functions for other road contexts, e.g. intersections scenario types, has not been tackled yet. In this work, we transfer existing highway fitness function templates to intersections, provide assistance on how to avoid possible pitfalls when applying search technology for scenario-based testing, and demonstrate the applicability of these transferred templates.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Fitness Function Templates for Testing Automated and Autonomous Driving Systems in Intersection Scenarios


    Contributors:


    Publication date :

    2021-09-19


    Size :

    1141698 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English





    Trajectory Planning for Automated Driving in Intersection Scenarios using Driver Models

    Speidel, Oliver / Graf, Maximilian / Kaushik, Ankit et al. | ArXiv | 2020

    Free access

    Trajectory Planning for Automated Driving in Intersection Scenarios Using Driver Models

    Speidel, Oliver / Graf, Maximilian / Kaushik, Ankit et al. | IEEE | 2020


    Did We Test All Scenarios for Automated and Autonomous Driving Systems?

    Hauer, Florian / Schmidt, Tabea / Holzmuller, Bernd et al. | IEEE | 2019