【摘要】 机器人自主移动导航是近年来研究的热点。针对蚁群优化(ACO)算法存在收敛速度慢以及易陷入局部最优的问题,提出了一种改进的ACO算法来解决机器人路径规划问题。上述算法将改进的人工势场(APF)算法和蚁群算法相结合,采用改进APF算法进行初始地图规划,减少了ACO算法初始规划的盲目性。算法利用A*算法的评估函数以及路径转折角度来改进启发函数,引入启发信息递增函数,免于局部最优的同时保证收敛速度。改进算法的信息素更新机制和路径评价函数,提高了算法的全局最优性,使得到的路径更符合实际需求。通过改进该算法的信息素更新机制和路径评价函数,提高了算法的全局最优性,得到的路径更符合实际需求。仿真结果表明,改进算法能提升收敛速度和最优解。更多还原

【Abstract】 Autonomous mobile navigation of robots has been a hot research topic in recent years. For the problem that the ant colony optimization(ACO) algorithm has slow convergence speed and easy to be trapped into local optimum in path planning, an improved ACO algorithm was proposed to solve the problem of robot path planning by combining an improved artificial potential field(APF) algorithm and the ACO algorithm. The improved APF algorithm was used to plan the initial map, which reduced the blindness of the ACO algorithm. By using the evaluation function of the A* algorithm and path turning angle to improve the heuristic function, and introducing the increasing function of heuristic information, the proposed algorithm avoided the local optimization and ensured the convergence rate of the algorithm. By improving the pheromone updating mechanism and path evaluation function of the algorithm, the global optimality of the algorithm was increased and the obtained path was more consistent with the actual requirements. Moreover, the improved state transition rule would adaptively change the selection probability to select the state transition function, increase the diversity of solutions and enhance the efficiency of the algorithm, so that the optimal solution would be found more quickly. Simulation results show that the proposed algorithm can improve the convergence speed and the optimal solution.更多还原