改进蚁群算法的路径规划研究

Study on path planning of improved ant colony algorithm

针对蚁群算法在复杂环境下收敛速度慢且存在停滞问题,提出一种改进的蚁群算法。为了避免蚁群陷入死锁状态,采用回退策略,避免蚂蚁盲目搜索产生大量交叉路径并有效减少蚂蚁死亡数量,并且借鉴了狼群分配策略来更新信息素,提高算法全局性,在状态转移概率中引入一个启发因子并进行调整,避免算法陷入停滞。仿真实验结果表明,改进后的蚁群算法收敛速度明显加快,寻优最短路径达到29.73,迭代次数较少28。验证了该算法的有效性和可行性。

Aiming at the problem of slow convergence speed and stagnation of ant colony algorithm in complex environment,An improved ant colony algorithm is proposed.To avoid the colony getting stuck in a deadlock,Use a fallback strategy,Avoid ant blind search to produce a large number of cross paths and effectively reduce the number of ant deaths,In addition,we also use the Wolf pack allocation strategy to update the pheromone and improve the overall performance of the algorithm.An heuristic factor is introduced into the state transition probability and adjusted to avoid the stagnation of the algorithm,The optimal shortest path reaches 29.73,and the number of iterations is less than 28.Simulation results show that the convergence speed of the improved ant colony algorithm is obviously accelerated,which verifies the effectiveness and feasibility of the algorithm.