Study of Ant Colony Optimization Algorithm Based on Adaptive Theory

Ant Colony Optimization (ACO)Algorithm is a new optimization algorithm. Preliminary study has shown that it has many promising futures. It provides a possible way for complicated combinatorial optimization problems. But it has the limitation of stagnation. In this paper, I take advantage of ACO to search automatically the optimal parameters In the solution space, that is through the dynamic path selection policy and global information update paralleling with local information update to adjust the ants search direction, and ultimately find the optimal parameters. the algorithm in this paper can improve the overall solution under the conditions of guaranteed convergence speed. Through the application of TSP problems the performance of ACO is optimized by the adaptive variation of the parameter in the algorithm. It shows that the improved algorithm can find better paths at higher convergence speed.