The invention provides a decision-making method and system for an unmanned vehicle. The method comprises the following steps: acquiring a first moment observation state in an intersection environment; inputting a time sequence observation state expanded from the first moment observation state into a trained first neural network to output a deterministic action of the target vehicle; receiving a time sequence observation state, inputting prediction information into a Markov model, modifying a full-connection neural network in a depth deterministic strategy gradient algorithm into an LSTM neural network to form a decision model based on a time sequence, and calculating a reward value of the decision model after a deterministic action; and obtaining a second moment observation state to jointly form prediction information, updating the time sequence observation state, re-executing the decision model, and updating the reward value until the target vehicle passes through the intersection environment. Based on the method, the invention further provides a decision making system. According to the method, the meta-exploration learning algorithm and the time double-delay depth deterministic strategy gradient algorithm are fused, and the vehicle passing safety and efficiency are effectively improved.
本发明提出了一种无人驾驶车辆的决策方法和系统,该方法包括获取交叉口环境下第一时刻观测状态;将第一时刻观测状态扩展出的时序观测状态输入至训练完成的第一神经网络中输出目标车辆的确定性动作;接收时序观测状态,并将预测信息输入马尔科夫模型中,以及将深度确定性策略梯度算法中的全连接神经网络修改为LSTM神经网络形成基于时序的决策模型,计算确定性动作后决策模型的奖励值;获取第二时刻观测状态共同形成预测信息,更新时序观测状态,重新执行决策模型,更新奖励值直到目标车辆穿越交叉口环境。基于该方法,还提出了决策系统。本发明融合元探索学习算法与时间双延迟深度确定性策略梯度算法,有效提升车辆通行安全和效率。
Decision-making method and system for unmanned vehicle
一种无人驾驶车辆的决策方法和系统
2022-07-22
Patent
Electronic Resource
Chinese
IPC: | G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / 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 |
European Patent Office | 2021
|Unmanned vehicle transverse dynamic avoidance decision-making method and unmanned vehicle system
European Patent Office | 2025
|Unmanned logistics vehicle behavior decision-making method and system
European Patent Office | 2023
|