The invention discloses an urban road automatic parking method based on reinforcement learning, and the method comprises the steps: S1, a parking space line recognition process, S2, a state input process, and S3, a parking decision control process, and the parking decision control process comprises the steps: S31, enabling Actor to predict a current state value through employing a calculated state value in an experience pool and a Critic network, and carrying out the updating of a neural network, taking the M mean square error as a Loss function, training the neural network, and obtaining an updated target neural network; and S32, inputting the state input vector into the updated target neural network corresponding to the parking action strategy pi *, outputting the obtained steering angle and acceleration of the vehicle, and controlling the movement of the vehicle.
本发明公开了一种基于强化学习的城市道路自动泊车方法,包括:S1、车位线识别流程,S2、状态输入流程,以及S3、泊车决策控制流程,其中,泊车决策控制流程包括:S31、Actor使用经验池中计算好的状态价值与Critic网络预测当前状态价值进行神经网络的更新,用M均方差误差作为Loss函数,对神经网络进行训练,得到更新后的目标神经网络;S32、将状态输入向量输入停车动作策略π*对应的更新后的目标神经网络,输出得到车辆的转向角和加速度,控制车辆的移动。
Urban road automatic parking method and system based on reinforcement learning
一种基于强化学习的城市道路自动泊车方法及系统
2022-05-10
Patent
Electronic Resource
Chinese
IPC: | 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 / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen |
Automatic parking method based on deep reinforcement learning
European Patent Office | 2020
|Automatic parking system and method based on urban parking platform
European Patent Office | 2024
|Airport guide vehicle automatic parking method based on reinforcement learning
European Patent Office | 2025
|Urban road parking space identification method and system based on deep learning
European Patent Office | 2022
|Full-automatic parking lot scheduling method based on deep reinforcement learning
European Patent Office | 2020
|