This paper proposes a methodology combining Long-Short-Term-Memory (LSTM)-assisted kinematic motion prediction with a hybrid coding algorithm for compressing the trajectory data of Connected Autonomous Vehicles (CAVs). The vehicle locations after the first two time steps are predicted based on the vehicle positions at the first two time steps and the kinematic equation. The vehicle velocities and accelerations are predicted based on the vehicle locations and LSTM. The hybrid coding algorithm integrates differential coding, Binary Coded Decimal (BCD) coding and arithmetic coding. Differential coding converts the original data into the difference between the original data and the predicted data. Since the length of the original data is large but the difference between it and predicted data is small, the required space for storing the data can be greatly reduced. BCD coding converts subsequences of different lengths to the subsequences with the same length so that the original information can be correctly reproduced after decompression. Arithmetic coding expresses the information in small space by converting the character sequence into a decimal between 0 and 1. The proposed algorithm is evaluated on the Next Generation Simulation Trajectory dataset. The experiment results show that the compression ratio and compression rate obtained by the proposed algorithm are respectively higher and lower than those obtained by the baseline algorithms. Also, the sum of compression time, decompression time and transmission time associated with the proposed algorithm is less than that associated with most baseline algorithms and transmission without compression.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Compressing Vehicle Trajectory Data Using Hybrid Coding With Kinematic Motion Prediction


    Contributors:
    Xu, Lipeng (author) / Wu, Zhizhou (author) / Wang, Yinhai (author) / Tang, Jinjun (author) / Liang, Yunyi (author)


    Publication date :

    2024-06-01


    Size :

    2595733 byte




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English



    Compressing trajectory data of a navigation device

    DUAN NING / WANG ZHI HU / JI PENG et al. | European Patent Office | 2017

    Free access

    Deep learning vehicle trajectory prediction method fused with kinematic model

    LI LINHUI / FU YIFAN / WANG XUECHENG et al. | European Patent Office | 2023

    Free access

    Vehicle vision system using kinematic model of vehicle motion

    TURK JAMES | European Patent Office | 2015

    Free access

    Vehicle Trajectory Prediction using GAN

    Hegde, Chinmayi / Dash, Suman / Agarwal, Pooja | IEEE | 2020


    A Hybrid Dynamic-Kinematic EKF for Train Trajectory Estimation

    Sessa, Pier Giuseppe / De Martinis, Valerio / Corman, Francesco | IEEE | 2018