One of the critical pieces of the self-driving puzzle is understanding the surroundings of a self-driving vehicle (SDV) and predicting how these surroundings will change in the near future. To address this task we propose MultiXNet, an end-to-end approach for detection and motion prediction based directly on lidar sensor data. This approach builds on prior work by handling multiple classes of traffic actors, adding a jointly trained second-stage trajectory refinement step, and producing a multimodal probability distribution over future actor motion that includes both multiple discrete traffic behaviors and calibrated continuous position uncertainties. The method was evaluated on large-scale, real-world data collected by a fleet of SDV s in several cities, with the results indicating that it outperforms existing state-of-the-art approaches.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    MultiXNet: Multiclass Multistage Multimodal Motion Prediction


    Beteiligte:
    Djuric, Nemanja (Autor:in) / Cui, Henggang (Autor:in) / Su, Zhaoen (Autor:in) / Wu, Shangxuan (Autor:in) / Wang, Huahua (Autor:in) / Chou, Fang-Chieh (Autor:in) / Martin, Luisa San (Autor:in) / Feng, Song (Autor:in) / Hu, Rui (Autor:in) / Xu, Yang (Autor:in)


    Erscheinungsdatum :

    2021-07-11


    Format / Umfang :

    7256394 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch





    Welfare Effects of Congestion Pricing and Transit Services in Multiclass Multimodal Networks

    Liu, Yang / Nie, Yu (Marco) | Transportation Research Record | 2012


    Multimodal interaction-aware motion prediction for autonomous street crossing

    Radwan, Noha / Burgard, Wolfram / Valada, Abhinav | BASE | 2020

    Freier Zugriff

    Multistage autonomous vehicle motion planning

    BAGNELL JAMES ANDREW / JAVDANI SHERVIN / NARAYANAN VENKATRAMAN | Europäisches Patentamt | 2023

    Freier Zugriff