The invention discloses a full-automatic parking lot scheduling method based on deep reinforcement learning. A scheduling problem of a full-automatic parking lot is regarded as a global optimal cooperation problem under multi-agent cooperation. The method comprises the steps of 1, establishing a state model of a parking lot environment; 2, establishing a model of the agent, wherein the model comprises a state, an action and a return function; 3, initializing an experience pool; and 4, proposing a global-model based deep deterministic policy gradient algorithm GDDPG (Global-model based Deep Deterministic Policy Gradient), establishing and training a network model for each agent until the precision of the model reaches a preset threshold, so as to obtain an optimal scheme of multi-agent collaborative path planning. By establishing the state model of the parking lot, priori knowledge and constraints suitable for the full-automatic parking lot are described in a formalized mode; and the GDDPG network is trained based on a deep reinforcement learning algorithm. Through combination of the GDDPG network and the deep reinforcement learning algorithm, a parking path is planned for each agent, the global optimization target is met, and the intelligence of the full-automatic parking lot is achieved.
本发明公开一种基于深度强化学习的全自动停车场调度方法,将全自动停车场的调度问题视作多智能体协作下的全局最优协作问题。步骤包括:1.建立停车场环境的状态模型;2.建立智能体的模型,包括状态、动作和回报函数;3.初始化经验池;4.提出一种带全局模型的深度确定性策略梯度算法GDDPG(Global‑model based Deep Deterministic Policy Gradient),为每个智能体搭建网络模型并进行训练,直到模型精度达到预设阈值,得到多智能体协同路径规划的最优方案。本发明通过建立停车场的状态模型,形式化描述适用于全自动停车场的先验知识和约束;基于深度强化学习算法,训练GDDPG网络。二者结合,为每个智能体规划一条泊车路径,并满足全局最优化目标,实现了全自动停车场的智能性。
Full-automatic parking lot scheduling method based on deep reinforcement learning
一种基于深度强化学习的全自动停车场调度方法
2020-12-11
Patent
Electronic Resource
Chinese
IPC: | G05B Steuer- oder Regelsysteme allgemein , CONTROL OR REGULATING SYSTEMS IN GENERAL / 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 |
Automatic parking method based on deep reinforcement learning
European Patent Office | 2020
|Deep reinforcement learning method suitable for automatic parking of high-speed train
European Patent Office | 2022
|Train accurate parking control method based on deep reinforcement learning
European Patent Office | 2022
|Airport guide vehicle automatic parking method based on reinforcement learning
European Patent Office | 2025
|Train temporary parking strategy optimization method based on deep reinforcement learning
European Patent Office | 2023
|