Pedestrian trajectory prediction in dynamic, multi-agent traffic junction scene is an important problem in the context of self-driving cars. Accurately predicting the trajectory of the agent, especially the pedestrian with high randomness, is of great significance to autonomous driving technology. In this paper, we propose Multi-PPTP, a novel trajectory prediction model that first utilizes a composite rasterized map to model the complex representations and interactions of road components, including dynamic obstacles (e.g., pedestrians, vehicles, and cyclists) and static road information (e.g., lanes, traffic lights, and sidewalks). Then it uses MapNet and AgentNet to extract spatio-temporal features by deep convolutional networks and LSTM to automatically derive relevant features, and next exploits Interaction-AttNet aggregate features to learn the high interactions among all components by affine transformation and multi-head attention mechanism. Additionally, we also propose a series of unique loss functions to predict multiple possible trajectories of each pedestrian while estimating their probabilities. Following extensive offline evaluation and comparison to the state-of-the-art models, our approach outperforms significantly the state-of-the-art models on our large scale in-house and public nuScenes dataset.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Multi-PPTP: Multiple Probabilistic Pedestrian Trajectory Prediction in the Complex Junction Scene


    Beteiligte:
    Li, Linhui (Autor:in) / Zhou, Bin (Autor:in) / Lian, Jing (Autor:in) / Wang, Xuecheng (Autor:in) / Zhou, Yafu (Autor:in)

    Erschienen in:

    Erscheinungsdatum :

    2022-08-01


    Format / Umfang :

    4433882 byte




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Multi-Relational Pedestrian Trajectory Prediction in Complex Scenes

    Peng, Wenshuo / Cui, Zhoujuan / Duan, Yiping et al. | IEEE | 2022


    Pedestrian Trajectory Prediction Combining Probabilistic Reasoning and Sequence Learning

    Li, Yang / Lu, Xiao-Yun / Wang, Jianqiang et al. | IEEE | 2020



    Trajectory prediction method and device for pedestrian of special work type in road scene

    HE TAO / CHEN XIAOLEI / LIAO WENLONG et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Multiple Goals Network for Pedestrian Trajectory Prediction in Autonomous Driving

    Chen, Weihuang / Zheng, Fang / Shi, Liushuai et al. | IEEE | 2022