The invention provides an expressway monitoring method and system based on a deep reinforcement learning algorithm, and the method comprises the steps: determining whether to start a variable speed limit control strategy in an upstream region of a certain road section or not according to a comparison result of traffic capacity and a preset threshold value; a DDQN intelligent agent based on a deep reinforcement learning algorithm is used for monitoring the traffic state of each road section in an expressway, and a real-time neural network and a target neural network are used for controlling the variable speed limit of vehicles in each road section in the expressway. And the experience samples in the memory pool are repeatedly trained for multiple times to obtain an optimal speed limit value action, so that an optimal variable speed limit control strategy is obtained, and finally, the DDQN intelligent agent displays the optimal speed limit value. According to the method, the speed difference between the vehicles, rear-end collision and other accidents can be effectively reduced, the highway passing pressure is relieved, the passing efficiency and passing safety of the vehicles are improved, and optimization of the variable speed limiting control effect is achieved.
本发明提供一种基于深度强化学习算法的高速公路监控方法及系统,该方法通过根据交通能力与预设阈值的比对结果决定是否在某一路段的上游区域启动可变限速控制策略,在可变限速控制策略启动后,利用基于深度强化学习算法的DDQN智能体对高速公路中各路段的交通状态进行监控,并采用实时神经网络和目标神经网络对高速公路中各路段车辆的可变限速进行控制,再通过对记忆池中的经验样本进行多次重复训练以获得最优限速值动作,从而得到最优可变限速控制策略,最后DDQN智能体将最优限速值进行展示。本发明不仅能有效减少车辆间的速度差和追尾等事故的发生,缓解高速公路通行压力,还提高车辆的通行效率和通行安全性,实现可变限速控制效果的优化。
Highway monitoring method and system based on deep reinforcement learning algorithm
一种基于深度强化学习算法的高速公路监控方法及系统
2022-12-23
Patent
Electronic Resource
Chinese
European Patent Office | 2022
|Highway On-Ramp Truck Platooning Based on Deep Reinforcement Learning
Transportation Research Record | 2024
|OVERTAKING MANEUVERS IN SIMULATED HIGHWAY DRIVING USING DEEP REINFORCEMENT LEARNING
British Library Conference Proceedings | 2018
|