关键词: 移动机器人, 路径规划, 蚁群算法, A*算法, 引导方向机制

Abstract:

To improve slow search efficiency and achieve real-time obstacle avoidance in traditional ant colony algorithms, an adaptive ant colony algorithm was proposed. A guidance direction mechanism was introduced to shorten the time of node selection. The A* algorithm's path-finding mechanism was introduced into the heuristic function to reduce the length and number of circles of the optimal path solution. The route planned by the traditional A* algorithm was used as the initial iteration data of the ant colony algorithm in global path planning, so as to solve the problem of slow initial convergence of the ant colony algorithm. The breadth first search mechanism of the traditional A* algorithm was introduced into the ant colony algorithm to address the issue of multiple iterations in the algorithm. In order to verify the superiority of this algorithm, two representative environmental models were adopted and compared with two traditional ant colony algorithms and genetic algorithms through comprehensive experiments. Comparative experiments demonstrate that the adaptive path-finding ant colony algorithm exhibits significant advantages over AS, ACS, and EAS algorithms in path planning, featuring faster convergence and superior path generation.

Key words: mobile robot, path planning, ant colony algorithm, A* algorithm, guidance direction mechanism