基于人工势场引导的改进RRT机器人路径规划算法

Improved RRT Robot Algorithm for Path Planning Based on Guidance of Artificial Potential Field

针对随机扩展树收敛速度慢、效率低的缺点,提出以人工势场引导节点向目标点逼近,并与改进的转换测试结合实现树扩展的自适应调控。采用人工势场算法建立采样节点的价值函数,使得随机扩展树不断向低代价空间扩展,当陷入局部极小值时,对RRT算法的采样策略进行调节、自适应地寻找逃离路径,使搜索过程快速跳出局部极小值。仿真实验表明,人工势场引导随机树渐进目标点,并与转换测试结合,提高了算法的搜索效率。

To solve the problem of slow convergence and low efficiency of RRT algorithm, it is proposed that the artificial potential field guides the node to approach the target point, and combined with the improved transition test to achieve adaptive control of tree expansion. The artificial potential field algorithm is used to establish the cost function of the sampling node, so that the random tree is continuously extended to the low-cost space. When falling into local minimum, the sampling strategy of the RRT algorithm is improved to adaptively seek the escape path so that the search process can be made to jump out of the attractive areas of local minimum point quickly. The simulation experiments show that the search efficiency of the algorithm is enhanced with the artificial potential field guiding the sampling nodes to the target one gradually and combined with the conversion test.