蚁群算法的改进及其在最优路径中的应用

Improved Ant Colony Algorithm and Its Application

为了克服蚁群算法在机器人路径规划过程中收敛速度慢、效率较低,容易陷入局部最优等缺点,提出了一种多步长改进蚁群算法,该算法的主要思想及思路是同时在概率公式中加入了拐点参数,使路径更加平滑,并且提出了新的信息素奖励惩罚机制.将文章改进的蚁群算法应用在30×30的栅格障碍环境中,让机器人找到一条从起点到终点的无碰撞的最短路径.实验仿真结果表明改进的蚁群算法与基本蚁群算法找到的最优路径相同,长度都是43.36,但是两者需要的迭代次数差别较大,改进蚁群算法仅在第8次迭代后就可以找到最优路径.而基本蚁群算法在迭代19次才会收敛.即文章提出的改进蚁群算法具有较高的搜索效率.

In order to overcome the shortcomings of ant colony algorithm in the process of robot path planning,such as slow convergence speed,low efficiency,and easy to fall into local optimality,a multi-step improved ant colony algorithm was proposed.The main idea of the algorithm was to simultaneously use the probability formula where inflection point parameters were added to make the path smoother,and a new pheromone reward and punishment mechanism was proposed.The improved ant colony algorithm in this article was applied to the grid obstacle environment,allowing the robot to find a shortest collision-free path from the start to the end.Experimental simulation results showed that the optimal path found by the improved ant colony algorithm was the same as the basic ant colony algorithm,with a length of 43.36,but the number of iterations required by the two was quite different.The optimal path could be found by the improved ant colony algorithm only in the 8th iteration while the basic ant colony algorithm needed 19 iterations.That was,the improved ant colony algorithm proposed in this paper had a higher search efficiency.