Abstract: Aiming at the problem of iterative stagnation in solving the traveling salesman problem with polymorphic ant colony algorithm, the improved ant colony algorithm is proposed. Combining with the predatory search algorithm, a new fusion ant colony algorithm is proposed. This fusion algorithm introduces the following mechanisms. In the artificial ants path-selecting stage, the snoop path is set to be a priority, a penalty factor is set for a non-snoop path, the pheromone is applied to both the snoop and non-snoop paths, and the mechanism is used to avoid the polymorphic ant colony algorithm’s stagnation. The predator search strategy is applied in the neighborhood of the optimal result of each round of artificial ant, and the optimal solution is selected by the competitive mechanism to update the pheromone. The research shows that the above mechanism can overcome the shortcomings of the polymorphic ant colony algorithm. The predatory search strategy can be used to guide the distribution of pheromone effectively and speed up the convergence of the algorithm to the optimal solution and improve the quality of the optimal solution. Simulation results for the TSP problem prove the effectiveness of the proposed algorithm.

Key words: polymorphic ant colony algorithm, predatory search, Traveling Salesman Problem（TSP）

摘要： 结合捕食搜索策略对多态蚁群算法进行改良。该算法引入以下机制：在人工蚁选择路径阶段，设置侦查素路径为优先，为非侦查素路径设置惩罚因子；利用权值在侦查素和非侦查素路径都施加信息素，通过该机制避免多态蚁群算法陷入停滞；在每轮人工蚁最优结果的邻域应用捕食搜索策略，并通过竞争机制选择最优解更新信息素。通过TSP的仿真实验结果表明，提出的融合算法可以有目的地指导信息素分布，加快算法向最优解的收敛速度及提高最优解质量，克服传统多态蚁群算法的缺陷。

关键词: 多态蚁群算法, 捕食搜索, 旅行商问题（TSP）