An Improved Ant Colony Algorithm Based on Distribution Estimation

In last two decades, Ant colony algorithm got extensive application in combinatorial optimization, function optimization and other fields. Ant colony algorithm is easy to fall into local optimum. A novel estimation of distribution algorithm by fusion improvement on ant colony algorithm and PBIL estimation of distribution algorithm is proposed. The algorithm introduce probability distribution model of PBIL algorithm to guide route choice, which greatly improves the faults that positive feedback mechanism of pheromone. Although the hybrid ant colony algorithm has achieved good results, this is just the preliminary attempt of distributed estimation algorithm combined with ant colony algorithm. Probability distribution model of other distribution estimation algorithm can also be used to guide the choice of ant colony optimal path.