The invention discloses a federated learning traffic flow prediction method based on unmanned aerial vehicle cooperation, which combines a ground vehicle and an unmanned aerial vehicle to promote information sharing and fusion, and comprises the following steps: firstly, the ground vehicle and the unmanned aerial vehicle perform data acquisition, and the ground vehicle is responsible for training a local model and updating model parameters; and then, the unmanned aerial vehicle group adopts a global model aggregation method based on consensus confirmation to ensure the reliability and global applicability of the model. In the global aggregation process, comprehensively evaluating the credibility of the unmanned aerial vehicle by adopting a fuzzy membership evaluation method to ensure that only the trusted unmanned aerial vehicle participates in the model aggregation process; the fairness, rationality and excitation compatibility among the unmanned aerial vehicles participating in the federated learning process are realized by applying a reputation-based multi-dimensional contract theory method, so that the resource allocation of the unmanned aerial vehicles is optimized. The invention provides a fair excitation and robust model aggregation strategy, and meanwhile, the relation between energy consumption and training delay is balanced, so that the performance of traffic flow prediction is improved.

    本发明公开一种基于无人机协作的联邦学习交通流预测方法,联合地面车辆与无人机以促进信息共享和融合,首先地面车辆和无人机进行数据采集,地面车辆负责训练本地模型并更新模型参数,随后无人机群体采用基于共识确认的全局模型聚合方法,以确保模型的可靠性和全局适用性。在全局聚合过程中,采用模糊隶属度评价方法对无人机的可信度进行全面评估,以确保只有受信任的无人机参与模型聚合过程;运用基于信誉的多维契约理论方法来实现参与联邦学习过程的无人机之间的公平性、合理性和激励兼容性,从而优化无人机的资源配置。本发明提供了一种公平的激励和鲁棒的模型聚合策略,同时权衡能量消耗和训练延迟之间的关系,从而提高交通流量预测的性能。


    Access

    Download


    Export, share and cite



    Title :

    Federal learning traffic flow prediction method based on unmanned aerial vehicle cooperation


    Additional title:

    一种基于无人机协作的联邦学习交通流预测方法


    Contributors:
    ZHOU MAN (author)

    Publication date :

    2024-11-15


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    Unmanned aerial vehicle aerial video traffic flow statistical system

    HU HUAZHI / GUO RONG | European Patent Office | 2024

    Free access

    Unmanned Aerial Vehicle–Based Traffic Analysis

    Khan, Muhammad Arsalan | Online Contents | 2017


    Cooperation use method of unmanned vehicle and unmanned aerial vehicle nest

    LIANG DINGHAO / YU XUEHAI | European Patent Office | 2023

    Free access

    Traffic dispersion unmanned aerial vehicle

    LI DONG / LI MINGCONG / ZHANG PENGFEI et al. | European Patent Office | 2022

    Free access