有向外力场作用下机器人路径规划方法

针对有向外力场作用下基于格点的机器人路径规划方法不能得到最优解的问题,采用了基于水平集方法的路径规划方法.将机器人的粒子跟踪转变为曲线的数值演化,通过解哈密尔顿-雅克比方程得到最优时间路径,改进后向路径追踪的数值计算精度保证了规划路径的可行性.仿真结果表明,算法可以有效应对复杂的有向外力场,并且在强外力场中仍保持路径的可行性.采用的连续路径规划方法可以突破传统机器人路径规划算法基于格点搜索的限制,并有效利用空间中存在的外力场.

Path planning method for robot in directional external force field

Aiming at the problem that the grid-based path planning method for a robot in the external force field can not obtain the optimal solution, a path planning method based on the level set was proposed. The particle tracking of method was changed into the numerical propagation of curves, and the optimal time path was obtained through solving the Hamilton-Jacobi equation. In addition, the feasibility of planning path was guaranteed through improving the numerical accuracy of backward path tracking. The simulation results show that the algorithm can effectively deal with the complex directional force field, and still keep the feasibility of path in the strong external force field. The adopted continuous path planning method can break through the limitation of traditional grid-based path planning algorithm for the robot, and effectively utilize the external force field existing in the space.