多因素改进势场蚁群算法的路径规划

Path planning for multi-factor improved potential field ant colony algorithm

针对路径规划算法中蚁群算法对目标点盲目性较大且无法应对多路况等问题,提出了一种多因素改进势场蚁群算法.首先,算法引入人工势场法重新构造路径长度启发函数并加入势场力递减系数,从而解决蚁群算法迭代时间长且易陷入局部最优解的问题;然后,综合考虑了势场路径长度因子,路径平缓性因子以及平滑性因子,构建新的多因子启发式函数,以适应复杂多变的路面环境;最后,运用动态切点法对路径进行平滑处理,提高路径整体质量.仿真实验表明,该算法在复杂颠簸路面情况下具有较好的适应力,能够有效解决机器人路径规划问题.

Aiming at the problem that the ant colony algorithm in the path planning algorithm is blind to the target point and cannot cope with multi-path conditions,a multi-factor improved potential field ant colony algorithm is proposed.Firstly,the algorithm introduces the artificial potential field method to reconstruct the path length heuristic function and adds the potential field force decreasing coefficient,so as to solve the problem that the ant colony algorithm has a long iteration time and is easy to fall into the local optimal solution.Then,a new multi-factor heuristic function is constructed by comprehensively considering the path length factor,path flatness factor and smoothness factor of the potential field to adapt to the complex and changeable road environment.Finally,the dynamic tangent point method is used to smooth the path to improve the overall quality of the path.Simulation experiments show that the algorithm has good adaptability in complex bumpy road conditions,and can effectively solve the problem of robot path planning.