In order to promote the particles’ autonomic learning ability in the dynamic multi-swarm particle swarm optimization algorithm, a multi-swarm particle swarm optimization algorithm based on autonomic learning and elite swarm is proposed. According to the concept of autonomic learning , the algorithm can achieve the purpose of optimization through replacing the regrouping operation with choosing learning objects autonomously in base swarms and conducting local searches in elite swarms. The proposed algorithm is applied to six text functions and compared with the dynamic multi-swarm particle swarm optimization algorithm. The results show that the proposed algorithm has a good performance in the convergence speed, the search accuracy, the time consumption and so on.