摘要
针对传统人工势场法中存在的一些局部极小点问题,文中提出了一种基于人工力场的移动机器人路径规划方法。该方法将机器人与目标的相对距离引入斥力函数,定义斥力的一个分力方向与障碍物的影响范围相切,另一个分力方向与引力方向一致,规定斥力向量与引力向量的内积始终大于等于零。这样,克服了传统人工势场法中存在的一些局部极小点问题,尤其是三个典型的、具有代表性的局部极小点问题。MATLAB仿真结果表明该人工力场法可以避开上述局部极小点,有效的规划出从起点到目标点的无碰路径。
The paper presents an Artificial Force Functions （AFF） approach for mobile robot path planning to overcome some typical problems that may arise with the conventional Artificial Potential Functions（APF）. This method introduces the relative position between the robot and the goal into repulsive function. The repulsive function is composed of two parts ： one＇ s direction is tangency with the control area of the obstacle and the angle between gravitation and repulsive is less than 90 degree while the other＇ s direction is consistent with the gravitation. In doing so, the AFF attempts to overcome some typical problems that may arise with the conventional APF. Specifically, it can avoid local minima for three representative scenarios. Simulation results of the example with MATLAB show that the proposed mehtod can effectively construct a path - planning system with the capability of reaching a goal and avoiding obstacles.
出处
《计算机仿真》
CSCD
2007年第11期144-146,197,共4页
Computer Simulation
基金
内蒙古工业大学重点研究项目(ZD200609)
关键词
路径规划
人工势场函数
局部极小点
人工力场函数
Path - planning
Artificial potential functions
Local minima
Artificial force functions
