In recent years, great efforts have been devoted to deep imitation learning for autonomous driving control, where raw sensory inputs are directly mapped to control actions. However, navigating through densely populated intersections remains a challenging task due to uncertainty caused by uncertain traffic participants. We focus on autonomous navigation at crowded intersections that require interaction with pedestrians. A multi-task conditional imitation learning framework is proposed to adapt both lateral and longitudinal control tasks for safe and efficient interaction. A new benchmark called IntersectNav is developed and human demonstrations are provided. Empirical results show that the proposed method can achieve a success rate gain of up to 30% compared to the state-of-the-art.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Learning Autonomous Control Policy for Intersection Navigation With Pedestrian Interaction


    Beteiligte:
    Zhu, Zeyu (Autor:in) / Zhao, Huijing (Autor:in)

    Erschienen in:

    Erscheinungsdatum :

    2023-05-01


    Format / Umfang :

    14598380 byte




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    PEDESTRIAN AUTONOMOUS NAVIGATION SYSTEM

    SASAKI HIROYUKI / FUKUI MEGUMI / SAKANASHI TOMOJI et al. | Europäisches Patentamt | 2022

    Freier Zugriff

    Pedestrian Autonomous Positioning System Based on Inertial Navigation

    Xing, Bowen / Jia, Haonan / Liu, Pengfei et al. | British Library Conference Proceedings | 2020


    PEDESTRIAN NAVIGATION SYSTEM, AND PEDESTRIAN NAVIGATION PROGRAM

    OGAWA KOJI | Europäisches Patentamt | 2021

    Freier Zugriff

    An online approach for intersection navigation of autonomous vehicle

    Bai, Yang / Chong, Zhuang Jie / Ang, Marcelo H. et al. | IEEE | 2014


    Autonomous vehicle navigation and pedestrian detection in urban environments

    Macek, Kristijan / Spinello, Luciano / Triebel, Rudolph et al. | Tema Archiv | 2009