一种改进的SA-APF路径规划算法

An Improved SA-APF Path Planning Algorithm

针对基于模拟退火的人工势场算法存在局部最小值问题,引入一种随机逃逸力,提出一种改进的SA-APF(Simulated Annealing-Artificial Potential Field)算法,可使机器人快速逃离局部最优解。针对算法中仍存在目标不可达问题,根据模糊控制方法提出一种自适应斥力势场增益系数,在机器人遇到目标点附近有对称排列的障碍物时,使机器人能够到达目标点,并采用4次贝赛尔曲线拟合,使路径更光滑。仿真结果表明,在简单和复杂的障碍物的实验条件下,改进后的算法可以准确、高效地完成路径规划。

Aiming at the problem of local minimum value in the artificial potential field algorithm based on simulated annealing,this paper introduces a random escape force and proposes an improved SA-APF(Simulated Annealing-Artificial Potential Field)algorithm,which can enable the robot to quickly escape the local optimal solution.Aiming at the problem that the target is still unreachable in the algorithm,an adaptive repulsive potential field gain coefficient is proposed according to the fuzzy control method,which enables the robot to reach the target point when it encounters a symmetrically arranged obstacle near the target point,and uses 4 Bessel curve fittings to make the path smoother.Simulation results show that under the experimental conditions of simple and complex obstacles,the improved algorithm can accurately and efficiently complete the path planning.