The invention relates to a multi-vehicle system ramp collaborative decision control method and device based on iterative learning. The method comprises the following steps: obtaining object information of each target object in a target area; constructing a search space according to the object information and a decision tree structure; the search space comprises candidate sequences of various vehicle cluster passing types; based on the guidance of a heuristic Monte Carlo tree algorithm and target merging motion data, searching a minimum time passing sequence in the candidate sequence; the target merging motion data comprises a collision-free merging trajectory of the optimal time cost of each target object; and controlling each target object to pass through the target area based on the minimum time passing sequence and the target merging motion data. By adopting the method, the passing efficiency of multi-vehicle coordination control can be improved.
本申请涉及一种基于迭代学习的多车系统匝道协同决策控制方法和装置。所述方法包括:获取目标区域中各目标对象的对象信息;根据对象信息按照决策树结构,构建搜索空间;搜索空间包括各种车辆集群通过类型的候选序列;基于启发式蒙特卡洛树算法和目标合并运动数据的指导,在候选序列中搜索最小时间通过序列;目标合并运动数据包括各目标对象最优时间成本的无碰撞合并轨迹;基于最小时间通过序列和目标合并运动数据控制各目标对象通过目标区域。采用本方法能够提高多车协调控制的通行效率。
Multi-vehicle system ramp collaborative decision control method and device based on iterative learning
基于迭代学习的多车系统匝道协同决策控制方法和装置
2024-10-29
Patent
Electronic Resource
Chinese
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |
European Patent Office | 2023
|Ramp driving decision control method, vehicle and storage medium
European Patent Office | 2023
|Intelligent vehicle highway ramp convergence decision-making method based on reinforcement learning
European Patent Office | 2023
|