摘要
为改善传统蚁群算法在路径规划中存在的规划路径实用性差、收敛速度慢、易陷入局部最优等问题,提出一种改进多步长蚁群算法.改进算法以移动机器人视野域内所有可直达节点作为下一步可选节点集,采用多步长移动方式以任意方向任意步长寻找下一节点,提高算法寻优效率和路径规划多样性;节点之间初始信息素依各节点与当前节点和目标节点连线的距离采取不均匀分布,降低蚁群在算法初期搜索的盲目性;通过路径长度增大优质路径与劣质路径的信息素更新差距,改进启发函数,提高算法收敛速度.仿真结果表明,改进算法规划路径具有长度短、路径平滑度高、步数少的优点,更符合移动机器人实际使用需求,收敛速度明显加快,路径规划效果提升显著.
Improved multi-step ant colony algorithm was proposed to solve the problems of traditional ant colony algorithm in path planning,such as poor practicability,slow convergence speed and local optimization.All the direct nodes in the field of view of the mobile robot for the improved algorithm were taken as the next optional node set,the multi-step moving method was used to find the next node in any direction and at any step length,and the optimization efficiency of the algorithm and the diversity of path planning was improved.The initial pheromones among nodes were unevenly distributed according to the distance between each node and the connecting line between current and target node,the blindness of ant colony search in the initial stage of the algorithm was reduced.By increasing the pheromone update gap between the high-quality path and the low-quality path through the path length,the heuristic function and the convergence speed of the algorithm was improved.The simulation results show that the improved algorithm has the advantages of short length,high smoothness and less steps,which are more in line with the actual needs of mobile robots.The convergence speed and the effect of path planning are significantly improved.
作者
马滕
茅健
MA Teng;MAO Jian(School of Mechanical and Automotive Engineering,Shanghai University of Engineering Science,Shanghai 201620,China)
出处
《上海工程技术大学学报》
CAS
2023年第3期255-262,共8页
Journal of Shanghai University of Engineering Science
关键词
路径规划
蚁群算法
多步长
路径平滑度
信息素
path planning
ant colony algorithm
multi-step
path smoothness
pheromone