The invention discloses a networked intelligent vehicle formation decision-making method based on nested graph reinforcement learning, and belongs to the technical field of Internet of Vehicles and automatic driving. The method comprises the following steps: S1, collecting state information of vehicles in a formation, and processing the state information to obtain an inter-formation nested graph and an inter-vehicle nested graph; s2, performing feature extraction on the inter-formation nested graph and the inter-vehicle nested graph by adopting a feature extraction network, and obtaining the action of each intelligent vehicle based on the extracted features; the feature extraction network comprises a graph attention layer, a full connection layer and an activation layer, and a multi-head attention mechanism is introduced to improve the feature extraction capability; and S3, optimizing the action by adopting a reward function to obtain an intelligent vehicle decision-making method. According to the method, full-integrated formation decision control under efficient communication can be realized in a high-dynamic, high-complexity and strong-randomness highway environment.
本发明公开了一种基于嵌套图强化学习的网联智能车辆编队决策方法,属于车联网和自动驾驶技术领域。方法包括:S1、采集编队内车辆状态信息,并对所述状态信息进行处理,得到编队间嵌套图以及车辆间嵌套图;S2、采用特征提取网络对所述编队间嵌套图以及所述车辆间嵌套图进行特征提取,基于提取的特征得到每辆智能车辆的动作;所述特征提取网络包括:图注意力层、全连接层以及激活层,并引入多头注意力机制提高特征提取能力;S3、对所述动作采用奖励函数进行优化,得到智能车辆决策方法。本发明能够在高动态、高复杂度、强随机性的高速公路环境中,实现高效通信下全集成化编队决策控制。
Networked vehicle formation decision-making method based on nested graph reinforcement learning
一种基于嵌套图强化学习的网联车辆编队决策方法
2024-05-28
Patent
Elektronische Ressource
Chinesisch
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 / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen |
Intelligent networked vehicle formation decision-making method
Europäisches Patentamt | 2023
|Automatic driving vehicle microscopic decision-making method based on reinforcement learning
Europäisches Patentamt | 2020
|Parked vehicle subtask unloading decision-making method based on reinforcement learning
Europäisches Patentamt | 2023
|Automatic driving vehicle overtaking decision-making method based on deep reinforcement learning
Europäisches Patentamt | 2022
|Intelligent vehicle highway ramp convergence decision-making method based on reinforcement learning
Europäisches Patentamt | 2023
|