Probabilistic Roadmap(PRM)is a commonly used path planning algorithm in the field of mobile robots.Aiming at the problems of uneven distribution of sampling points,low efficiency of road map construction and unsmooth path redundancy in traditional PRM,an improved PRM is proposed.The two-dimensional Sobol sequence is used to optimize the sampling strategy to ensure the global uniform distribution of sampling points,optimize the coverage area of sampling points,and improve the quality of sampling points.Secondly,the sampling points are classified by neighborhood and the connection constraint is applied to connect the sampling points in the adjacent domain,which reduces the size of the roadmap and improves the composition and search efficiency of the roadmap.Then,the node translation optimization algorithm is used to optimize the node position,so that the optimized path conforms to the optimal path in the actual space.Finally,the Bessel curve is used to smooth the path inflection point,so that the generated path is more in line with the actual motion constraints of the robot.A large number of simulation results show that the improved PRM can effectively improve the quality of the planning path and is less affected by the number of sampling points.Compared with the traditional PRM and other improved PRM,the proposed PRM has obvious advantages in path length,running time and success rate.
关键词
概率路线图/路径规划/路径优化策略/Sobol序列/贝塞尔曲线
Key words
probability roadmap/path planning/path optimization strategy/Sobol sequence/Bessel curve
维普
万方数据