In this paper, we compare different deep neural network approaches for motion prediction within a highway entrance scenario. The focus of our work lies on models that operate on limited history of data in order to fulfill the Markov property1 and be usable within an integrated prediction and motion planning framework for automated vehicles. We examine different model structures and feature combinations in order to find a model with a good tradeoff between accuracy and computational performance. We evaluate all models with standard metrics like the negative log-likelihood (NLL) and evaluate the performance of each model within a closed-loop simulation. We find a neural network only operating on spatial features of the current state to have the best closed-loop prediction performance, despite the NLL suggesting otherwise.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Deep neural networks for Markovian interactive scene prediction in highway scenarios


    Contributors:


    Publication date :

    2017-06-01


    Size :

    357823 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Scene-Extrapolation: Generating Interactive Traffic Scenarios

    Zipfl, Maximilian / Schutt, Barbara / Zollner, J. Marius | IEEE | 2024


    Deep Recurrent Neural Networks for Highway Traffic Flow Prediction at Toll Stations

    Papagiannis, Tasos / Ioannou, George / Alexandridis, Georgios et al. | IEEE | 2024


    Interactive scene prediction for automotive applications

    Lawitzky, Andreas / Althoff, Daniel / Passenberg, Christoph F. et al. | IEEE | 2013


    INTERACTIVE SCENE PREDICTION FOR AUTOMOTIVE APPLICATIONS

    Lawitzky, A. / Althoff, D. / Passenberg, C. et al. | British Library Conference Proceedings | 2013


    A probabilistic long term prediction approach for highway scenarios

    Schlechtriemen, Julian / Wedel, Andreas / Breuel, Gabi et al. | IEEE | 2014