The invention discloses a parked vehicle subtask unloading decision-making method based on reinforcement learning, and the method comprises the steps: receiving a task unloading request through a wireless access point AP, observing a parked vehicle in a parking lot, considering the parking stability of the parked vehicle, and constructing a vehicle candidate set; the AP inputs a candidate set vehicle observation state and task attributes at the current moment into a reinforcement learning model for training, subtask division and unloading decision making are carried out, the influence of the mobility of a parked vehicle on task unloading is judged, and if the task cannot be completed in time due to the fact that the vehicle leaves, the task is not completed in time; task migration is carried out to ensure timely completion of tasks. According to the method, the parking stability of the vehicles is considered, a candidate set is constructed by adopting a reinforcement learning algorithm to screen the parked vehicles, the influence of vehicle dynamics on an existing unloading decision is reduced, multiple vehicles can cooperatively complete sub-tasks, task unloading time delay is optimized, meanwhile, the sub-tasks which cannot be completed are processed in time by adopting task migration, and the unloading efficiency of the vehicles is improved. And the task failure rate is reduced.

    本发明公开了一种基于强化学习的停放车辆子任务卸载决策方法,通过无线接入点AP接收任务卸载请求,观测停车场内停放车辆,考虑停放车辆的停留稳定性,构建车辆候选集,AP将当前时刻的候选集车辆观测状态和任务属性输入到强化学习模型中进行训练,进行子任务的划分和卸载决策,再通过判断停放车辆的移动性对任务卸载的影响,如果因为车辆驶离导致任务无法及时完成,进行任务迁移保证任务及时完成。本发明的方法考虑车辆的停留稳定性,采用强化学习算法构建候选集对停放车辆进行筛选,降低车辆动态性对已有卸载决策的影响,使多个车辆能够协同完成子任务,优化任务卸载时延,同时采用任务迁移及时处理无法完成的子任务,降低任务失败率。


    Access

    Download


    Export, share and cite



    Title :

    Parked vehicle subtask unloading decision-making method based on reinforcement learning


    Additional title:

    基于强化学习的停放车辆子任务卸载决策方法


    Contributors:
    ZHANG KE (author) / DU YANXI (author) / CAI MENGYU (author)

    Publication date :

    2023-06-30


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G06Q Datenverarbeitungssysteme oder -verfahren, besonders angepasst an verwaltungstechnische, geschäftliche, finanzielle oder betriebswirtschaftliche Zwecke, sowie an geschäftsbezogene Überwachungs- oder Voraussagezwecke , DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS



    Balance-Oriented Task Unloading Optimizing Algorithm for Parked Vehicle Edge Computing

    Hu, Xi / Guo, Lijun / Yao, Zhaoyang et al. | Springer Verlag | 2022


    Balance-Oriented Task Unloading Optimizing Algorithm for Parked Vehicle Edge Computing

    Hu, Xi / Guo, Lijun / Yao, Zhaoyang et al. | TIBKAT | 2022


    Balance-Oriented Task Unloading Optimizing Algorithm for Parked Vehicle Edge Computing

    Hu, Xi / Guo, Lijun / Yao, Zhaoyang et al. | British Library Conference Proceedings | 2022


    PARKED VEHICLE SEARCH SYSTEM

    TANIGUCHI AKIRA | European Patent Office | 2021

    Free access

    Determining vehicle parked location

    KAMINI ADITYA S / ELNAJJAR HASSAN A / ELSWICK RICHARD | European Patent Office | 2020

    Free access