论文部分内容阅读
在自由空间下粒子群算法(Particle Swarm Optimization,PSO)求解机器人路径规划问题时,存在易早熟、编码维度高和路径不平滑等问题,为此提出了一种基于PSO和三次样条插值的路径规划方法。所设计的粒子编码为环境中若干个路径节点的坐标,路径节点的个数决定了样条曲线的个数同时也决定了路径转向的次数。通过3次样条函数对路径的起点、路径节点和终点进行插值,从而得到一条由插值点构成的路径。仿真结果表明,相比传统方法所提出的算法能快速找到平滑的最优路径,并且能为多个机器人规划出最优的无碰路径。
Particle swarm optimization (PSO) in free space has some problems such as precocious precoding, high coding dimension and path non-smoothness when it is used to solve robot path planning. Therefore, a path based on PSO and cubic spline interpolation Planning method. The designed particle code is the coordinates of several path nodes in the environment. The number of path nodes determines the number of spline curves and also determines the number of path steering. Through the cubic spline function, the starting point of the path, the path node and the end point are interpolated to obtain a path formed by the interpolation points. The simulation results show that the proposed algorithm can find the smoothed optimal path quickly compared with the traditional method and can plan the optimal path without collision for multiple robots.