计算机工程与应用 ›› 2018, Vol. 54 ›› Issue (6): 105-109.DOI: 10.3778/j.issn.1002-8331.1610-0131

• 模式识别与人工智能 • 上一篇    下一篇

基于模糊人工势场法的智能全向车路径规划

韩  伟,孙凯彪   

  1. 大连理工大学 控制科学与工程学院,辽宁 大连 116024
  • 出版日期:2018-03-15 发布日期:2018-04-03

Research on dynamic path planning of fuzzy artificial potential field method

HAN Wei, SUN Kaibiao   

  1. School of Control Science and Engineering, Dalian University of Technology, Dalian, Liaoning 116024, China
  • Online:2018-03-15 Published:2018-04-03

摘要: 针对于移动机器人在传统人工势场法路径规划中易于陷入局部最小点而无法抵达目标点的问题,同时考虑到实际环境中人工势场法相关参数的不确定性,提出了一种基于模糊人工势场法的动态路径规划方法。借助于专家经验进行模糊决策,调整移动机器人在各个时刻的合力大小和方向,进而解决斥力常数、引力方向偏角以及机器人行驶速度的不确定性问题。为了验证该方法的有效性,在智能全向车平台进行了应用,结果表明,智能全向车运动轨迹平滑,避免了实际应用中的震荡问题。

关键词: 人工势场法, 移动机器人, 路径规划, 模糊决策, 局部最小点

Abstract: Aiming at the problem faced in the traditional artificial potential field method that the robot cannot reach the target point due to the local minimum point in the path planning, a dynamic path planning method based on the fuzzy artificial potential field method is proposed by considering that the relevant parameters of the artificial potential field method are not constant in actual environment. The fuzzy decision is made by the expert’s experience, which is used to adjust the force magnitude and direction of the robot at each moment, and then deals with uncertainty of the repulsive force constant, the gravitational direction deviation angle and the robot speed. To verify the efficiency of the proposed method, it is applied to a four-wheel omni-directional robot platform and the results show that the motion trajectory of the robot is smoother and the oscillation problem is avoided successfully.

Key words: artificial potential field method, mobile robot, path planning, fuzzy decision, local minimum point