Accurate trajectory prediction is crucial for the safe and efficient operation of autonomous vehicles. The growing popularity of deep learning has led to the development of numerous methods for trajectory prediction. While deter-ministic deep learning models have been widely used, deep generative models have gained popularity as they learn data distributions from training data and account for trajectory uncertainties. In this study, we propose EquiDiff, a deep generative model for predicting future vehicle trajectories. EquiDiff is based on the conditional diffusion model, which generates future trajectories by incorporating historical information and random Gaussian noise. The backbone model of EquiDiff is an SO(2)-equivariant transformer that fully utilizes the geometric properties of location coordinates. In addition, we employ Recurrent Neural Networks and Graph Attention Networks to extract social interactions from historical trajectories. To evaluate the performance of EquiDiff, we conduct extensive experiments on the NGSIM dataset. Our results demonstrate that EquiDiff outperforms other baseline models in shortterm prediction, but has slightly higher errors for long-term prediction. Furthermore, we conduct an ablation study to investigate the contribution of each component of EquiDiff to the prediction accuracy. Additionally, we present a visualization of the generation process of our diffusion model, providing insights into the uncertainty of the prediction.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    EquiDiff: A Conditional Equivariant Diffusion Model For Trajectory Prediction


    Contributors:
    Chen, Kehua (author) / Chen, Xianda (author) / Yu, Zihan (author) / Zhu, Meixin (author) / Yang, Hai (author)


    Publication date :

    2023-09-24


    Size :

    1428112 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Conditional agent trajectory prediction

    MAHJOURIAN REZA / DOWNEY CARLTON MACDONALD / SAPP BENJAMIN et al. | European Patent Office | 2024

    Free access

    CONDITIONAL AGENT TRAJECTORY PREDICTION

    MAHJOURIAN REZA / DOWNEY CARLTON MACDONALD / SAPP BENJAMIN et al. | European Patent Office | 2022

    Free access

    RESET: Revisiting Trajectory Sets for Conditional Behavior Prediction

    Schmidt, Julian / Huissel, Pascal / Wiederer, Julian et al. | IEEE | 2023


    Conditional Wasserstein Auto-Encoder for Interactive Vehicle Trajectory Prediction

    Fei, Cong / He, Xiangkun / Kawahara, Sadahiro et al. | IEEE | 2020


    Vehicle Trajectory Prediction Using Intention-based Conditional Variational Autoencoder

    Feng, Xidong / Cen, Zhepeng / Hu, Jianming et al. | IEEE | 2019