This paper presents a path-planning strategy for autonomous vehicles which aims to provide safe and feasible manoeuvres in various driving environments. Our strategy uses a hierarchical architecture which consists of three components: a behaviour planner, a map and path selector and a local-path planner. The behaviour planner performs the rule-based decision process which determines the overall vehicle manoeuvres. The map and path selector preprocesses perception data and chooses a local-path-planning algorithm using the results of the behaviour planner. From this selection, the local-path planner generates a driveable and collision-free path. For reliable path generation under various driving conditions, the proposed local path planner employs two algorithms: a road-model-based path planning algorithm and a graph-structure-based path-planning algorithm. The former is used for structured road driving, such as lane keeping or changing, and the latter is used for unstructured road driving. The proposed hierarchical path-planning algorithm was implemented in the autonomous vehicle called A1, which was applied with an in-vehicle-network-based distributed system architecture. A1 won the 2012 Autonomous Vehicle Competition organized by the Hyundai Motor Group in Korea.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Road-model-based and graph-structure-based hierarchical path-planning approach for autonomous vehicles


    Beteiligte:
    Kim, Junsoo (Autor:in) / Jo, Kichun (Autor:in) / Chu, Keonyup (Autor:in) / Sunwoo, Myoungho (Autor:in)


    Erscheinungsdatum :

    01.07.2014


    Format / Umfang :

    20 pages




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch