The invention discloses an unmanned driving decision-making method and system based on a Monte Carlo search tree, and the method comprises the following steps: 1, carrying out the initialization, and constructing an N-layer tree structure; the current lane is used as a root node, the routing relation between lanes is incrementally constructed in the unmanned driving process through the nodes and the edge structure of the Monte Carlo search tree, and each tree node represents one lane; 2, simulation is carried out according to current perception information and task information. According to the method, the characteristics of discreteness, long lane sequence and few selectable lanes of each section of road of a lane-level decision-making task are fully utilized, an expansion mechanism of the Monte Carlo tree is combined, a Monte Carlo tree network structure is incrementally constructed, and the problem of high calculation complexity caused by an exhaustion method carried out by a long lane sequence is avoided. Meanwhile, human driving data and a deep neural network are utilized to construct a new scoring mechanism.
本发明公开了一种基于蒙特卡洛搜索树的无人驾驶决策方法及其系统,包括以下步骤:步骤一:初始化,构建N层树结构;以当前车道为根节点,利用蒙特卡洛搜索树的节点和边结构在无人驾驶过程中增量式地构建车道与车道之间的路由关系,每一个树节点代表一条车道;步骤二:模拟,根据现感知信息和任务信息。本发明充分利用车道级决策任务的离散性、车道序列长、每段道路可选车道少的特征,结合蒙特卡洛树的扩展机制,增量式地构建蒙特卡洛树网络结构,避免了因长车道序列进行穷举法带来的计算复杂度高的问题。同时利用人类驾驶数据和深度神经网络,构建一种新的评分机制。
Unmanned driving decision-making method and system based on Monte Carlo search tree
一种基于蒙特卡洛搜索树的无人驾驶决策方法及其系统
2024-04-02
Patent
Electronic Resource
Chinese
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 / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen |
Monte Carlo Tree Search for online decision making in smart industrial production
BASE | 2021
|European Patent Office | 2021
|MONTE CARLO TREE SEARCH METHODS FOR TELESCOPE TASKING
TIBKAT | 2020
|