Research on path optimization of ant colony algorithm Improved Particle Swarm Optimization and Reverse Learning

Abstract

Aiming at the difficulty of determining the key parameters when applying ant colony optimization algorithm (ACO) to traveling salesman problem, we propose an improved particle swarm optimization (PSO) algorithm for adaptive parameter acquisition.Because repeated calls to ACO will increase the cost of computing and get the local optimal solution easily, the number of single ACO iterations is reduced, and the update of the pheromone is determined by the fitness function.After each call to ACO, the pheromone is not adjusted.In order to get better quality parameters of PSO, the reverse learning strategy is applied to PSO, and the speed of optimization is improved.The effectiveness of the algorithm is proved by the simulation experiment.