The invention discloses a CAV decision-making method applying selective federated learning. The method comprises the steps that all automatic driving automobiles download global models of last iteration updating from an FL server; training a local model by all the automatic driving automobiles by using own driving data, and updating model parameters; acquiring communication overhead and calculation overhead of all the autonomous vehicles in the federated learning process; multi-factor driven local model screening is carried out, the selected autonomous vehicle uploads the updated local model to an FL server, and local model screening comprises screening based on reputation, screening based on model convergence and screening based on maximized utility; the FL server aggregates the received local models to obtain an improved global model. According to the method, a multi-factor driven model screening strategy is provided before the local model is uploaded, the defect that participants are not distinguished in a traditional method is overcome, and high-quality users can be selected as many as possible.

    本发明公开了一种应用选择性联邦学习的CAV决策方法,包括:所有自动驾驶汽车从FL服务器下载上一轮迭代更新的全局模型;所有自动驾驶汽车使用自身的行驶数据训练本地模型,更新模型参数;获取所有自动驾驶汽车在联邦学习过程中的通信开销和计算开销;进行多因素驱动的本地模型筛选,选中的自动驾驶汽车将更新后的本地模型上传至FL服务器,其中,本地模型筛选包括基于信誉的筛选、基于模型收敛性的筛选和基于最大化效用的筛选;所述FL服务器对接收到的本地模型进行聚合以获得改进的全局模型。本发明在上传本地模型之前提出了一种多因素驱动的模型筛选策略,克服传统方法不对参与者加以区分的缺陷,能够选择尽可能多的高质量用户。


    Access

    Download


    Export, share and cite



    Title :

    CAV decision-making method applying selective federated learning


    Additional title:

    一种应用选择性联邦学习的CAV决策方法


    Contributors:
    FU YUCHUAN (author) / TANG XINLONG (author) / LI CHANGLE (author)

    Publication date :

    2023-01-31


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion



    Decision Making in Action: Applying Research to Practice

    Orasanu, Judith / Statler, Irving C. | NTRS | 1994


    Decision Making in Action: Applying Research to Practice

    Orasanu, Judith / Hart, Sandra G. | NTRS | 1994


    A Selective Federated Reinforcement Learning Strategy for Autonomous Driving

    Fu, Yuchuan / Li, Changle / Yu, F. Richard et al. | IEEE | 2023