The invention discloses a traffic flow path planning trajectory selection method based on reinforcement learning. The method comprises the following steps: S1, traffic flow simulation environment modeling; s2, collecting online data; s3, traffic flow trajectory planning decision making; wherein the traffic flow trajectory planning decision comprises behavior decision and trajectory planning, the behavior decision outputs vehicle decision actions according to vehicle surrounding environment information, behaviors such as lane changing, deviation, acceleration and deceleration and the like, and the trajectory planning determines a target state according to the decision actions so as to plan a travelable trajectory in the transverse and longitudinal directions. According to the method, artificial intelligence and a traditional method are combined, the traffic flow trajectory planning and selecting method based on reinforcement learning is invented, the human driving intention is modeled through the neural network, and the travelable trajectory is generated according to the driving intention through the traditional method.
本发明公开了一种基于强化学习的交通流路径规划轨迹选择方法,包括如下步骤:S1、交通流仿真环境建模;S2、在线数据收集;S3、交通流轨迹规划决策;其中,交通流轨迹规划决策包括行为决策与轨迹规划,行为决策根据车辆周围环境信息输出车辆决策动作,有变道、偏移、加减速等行为,轨迹规划根据决策动作确定目标状态以在横纵方向规划出可行驶轨迹。本申请采用人工智能与传统方法相结合,发明一种基于强化学习的交通流轨迹规划选择的方法,通过神经网络建模人类驾驶意图,而采用传统方法根据驾驶意图生成可行使的轨迹。
Traffic flow path planning trajectory selection method based on reinforcement learning
一种基于强化学习的交通流路径规划轨迹选择方法
2024-10-29
Patent
Electronic Resource
Chinese
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / G01C Messen von Entfernungen, Höhen, Neigungen oder Richtungen , MEASURING DISTANCES, LEVELS OR BEARINGS / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen |
Crossroad lighthouse trajectory planning method based on deep reinforcement learning
European Patent Office | 2024
|Intention-driven reinforcement learning-based path planning method
European Patent Office | 2024
|BASE | 2020
|Parking path planning method based on reinforcement learning
European Patent Office | 2020
|INTENTION-DRIVEN REINFORCEMENT LEARNING-BASED PATH PLANNING METHOD
European Patent Office | 2024
|