中文摘要：

传统的快速扩展随机树（RRT）算法虽然有很多优良特性,但是由于扩展点的随机选取,规划出来的路径具有很大的随机性。文中在对RRT算法改进的基础上,提出了一种动态步长的RRT路径规划算法。其中步长为RRT生长的最小单位长度。动态步长的RRT算法是在对传统RRT算法的基础上,添加了动态步长的特性,改善了快速扩展随机树的不确定性,提高了避障能力,使得算法确定性和高避障能力兼备。仿真实验结果表明,该算法在路径规划中具有路径确定、速度快和高避障能力的特点。

英文摘要：

Although the traditional Rapidly-exploring Random Tree （RRT） algorithm has many good features, there is a lot of randomness in path planning of RRT because of the random selection of the vertex. Based on the improvement of RRT algorithm, a new RRT path planning algorithm of dynamic step size is proposed in this paper. The step size is the minimum unit length when RRT exploring. Based on traditional RRT, the dynamic step size is added, avoiding the uncertainty, and the obstacle avoidance capability is improved, thus the path planning of RRT algorithm has both obstacle avoidance ability and high certainty. The results of simulation experiments show that the algorithm has the features of avoiding the uncertainty, fast speed and obstacle avoidance in path planning.