为解决机器人在多障碍区的路径规划问题,提出了一种基于由遗传算法改进的人工势场法(APFIGA)的全方位移动操作机器人路径规划方法.该规划方法通过在可穿越的障碍区添加障碍物穿越系数来鼓励机器人走捷径,进而提高其路径优化能力；利用当前点邻域内势场强度信息,结合遗传算法确定全方位移动操作机器人的运动方向及速度,并引入机器人速度及移动障碍物速度的影响,得出势场强度下降最快的路径,提高方法的动态环境规划能力.该方法能克服机器人在障碍物附近易于抖动的现象,通过在斥力势中添加与目标距离成正比的系数项解决目标不可达问题,并利用填平势场跳出局部极小.仿真与实物实验结果验证了所提出路径规划方法的正确性和有效性.%A method for the path planning for an omni-directional mobile manipulator based on the artificial potential field method improved by the genetic algorithm (APFIGA) is proposed to solve the difficuties in robot path planning in a multiple obstacle area. The method uses the traversing coefficient to encourage a robot to traverse a traversable multiple obstacle area. Thus the path planning ability can be improved. And it can obtain a path in which the potential field strength decreases fast by using the manipulator' s neighborhood potential field strength information, and the movement direction and the speed of the manipulator is determined by using the genetic algorithm. Due to the introducing of the effect of robot speed and the mobile obstacle speed, the path planning ability under dynamic environments is improved. The problem that the robot is easy to shake around obstacles is solved. The target point un-reachable problem is solved by adding the coefficient item proportional to target distance to the repulsion potential. The local minimum value problem is solved by adding the additional potential field to fill local minimum. The results of simulation and physical experiment verified that the proposed path planning method was correct and effective.
展开▼
