改进A^(*)算法的移动机器人的路径规划

Improved A^(*)algorithm in path planning of mobile robot

针对传统A^(*)算法计算量大、搜索效率低、转折次数多等问题,提出了一种改进A^(*)算法。通过划分传统A^(*)算法的邻域节点等级,优先搜索高级组中的子节点,将双向平滑理念引入到Floyd-Warshall算法中,减少路径中的拐点个数,使规划路径更加平滑,对比仿真研究传统A^(*)算法和改进A^(*)算法。结果表明,改进后的A^(*)算法路径长度最大缩减2.8%,搜索时间及转折次数仅为传统A^(*)算法的58.94%和62.5%,寻优能力较好。

This paper proposes an improved A^(*)algorithm designed to eliminate drawbacks inherent in traditional A^(*)algorithm,such as larger amount of calculation,lower search efficiency,and more turning times.The study includes classifying the neighborhood nodes of the traditional A^(*)algorithm;preferentially searching for the child nodes in the advanced group;introducing the two-way smoothing concept into the Floyd-Warshall algorithm to reduce the number of inflection points in the path and making the planned path smoother;and simulating.A^(*)algorithm and improved A^(*)algorithm.The results show that the improved A^(*)algorithm demonstrates an better optimization ability thanks to the maximum reduction in the path length by 2.8%and the equivalence of the search time and the number of turns to 58.94%and 62.5%of the traditional A^(*)algorithm.