In this research, a procedure based on GRU method has been proposed to predict the short-term trajectory of drones. First, historical trajectory data of various types of drones have been collected. Then, the GRU neural network has been constructed for data training. Several benchmark methods are selected to testify the prediction performance of the present method. The results indicated that the proposed approach achieved generally satisfactory performance over most of the scenarios of the testing datasets. Moreover, both the one-step and multi-step trajectory predictions are conducted. Comparison of MAEs for different speed sections are also provided. The results can provide technical support for UAV safety management in low altitudes.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Short-term Trajectory Prediction for Small Scale Drones at Low-attitude Airspace


    Contributors:
    Zhu, Renwei (author) / Yang, Zhao (author) / Chen, Jun (author) / Li, Na (author) / Zhang, Zhijie (author) / Song, Yilu (author)


    Publication date :

    2022-10-12


    Size :

    1639046 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English




    An Airport Airspace Flow Control Method for Drones

    Zeng, Guoqi / Cui, Kai / Quan, Quan et al. | IEEE | 2019



    Short-term trajectory prediction methods

    Chatterji, Gano | AIAA | 1999


    Short-Term Trajectory Prediction Methods

    Chatterji, G. / AIAA | British Library Conference Proceedings | 1999