非结构化环境中基于拓扑约束的地面无人驾驶路径规划算法  

Path planning algorithm for ground unmanned vehicles based on topological constraints in unstructured environments

作　　者：刘东升[1] 王俊生 LIU Dong-sheng;WANG Jun-sheng(College of Information Science and Engineering,Northeastern University,Shenyang 110004,China)

机构地区：[1]东北大学信息科学与工程学院,沈阳110004

出　　处：《控制与决策》2022年第12期3103-3114,共12页Control and Decision

基　　金：中央高校基本科研业务费专项资金项目(N2124002-12);国家重点研发计划项目(2020YFE0201100);国家自然科学基金项目(61703429)。

摘　　要：针对非结构化环境地面无人驾驶路径规划过程中路径避障以及多车路径冲突的难题,通过同调以及de Rham上同调对环境中障碍物拓扑信息的精确描述,提出一种拓扑约束下基于A*算法且用时更短的路径规划算法.该算法可实现非结构化环境中多无人车全局路径的拓扑分类,从而为多车的协同规划提供一种新的研究思路.此外,结合C-空间动态广义Voronoi图(GVD)的路径拓扑分离特性,提出一种拓扑约束下可用于多无人车全局路径规划的高效算法-–C-空间-GVD-h_(S)增广A*算法.最后,通过Gazebo仿真平台模拟的具有多障碍物以及狭窄区域的非结构化环境,验证所提出方法的有效性以及与现有方法相比的优越性.In allusion to the research problems of path avoidance and multi-vehicles path conflicts during the path planning process of ground unmanned driving in unstructured environments,a shorter time-consuming path planning algorithm based on the A*algorithm under topology constraints is proposed by using homology and de Rham cohomology to accurately describe the topological information of obstacles.The algorithm realizes the topological classification of the global paths of multi-vehicles in unstructured environments,thereby providing a new research idea for the collaborative planning of multiple ground unmanned vehicles.Furthermore,combined with the topological separation characteristics of paths on the C-space dynamic generalized Voronoi diagram (GVD),an efficient algorithm called C-space-GVD-h_(S) augmented A*algorithm for global path planning of multi-vehicles under topological constraints is proposed.Finally,the unstructured environment with multiple obstacles and narrow areas simulated by the Gazebo simulation platform verifies the effectiveness of the proposed method and its superiority compared with the existing methods.

关 键 词：路径规划 de Rham上同调 拓扑约束 广义Voronoi图 C-空间-GVD-hS增广A*算法 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]