Author(s)

Zhaohua Long, Ruifang Dong

Corresponding Author

Zhaohua Long

ABSTRACT

Ant colony algorithm is based on Ant System (AS) and it is a very important group intelligence algorithm, which is used in many fields, but there are also some shortcomings. The classical ant colony algorithm is analyzed and studied, and an improved P-ACS algorithm is proposed based on ACS algorithm in this paper. Through analysis and experiment, it is found that although the performance of ACS algorithm is higher than AS algorithm, there are still some problems, such as: falling into local optimal solution, search stagnation, and slow initial convergence.The important reason for the above problems is that the pheromone update can not accurately reflect the actual situation of the path. Aiming at this problem, a P-ACS ant colony algorithm is proposed based on particle swarm optimization algorithm(PSO). The algorithm optimizes the pheromone update strategy from three aspects: pheromone concentration range setting, initial pheromone setting and global update strategy improvement.

KEYWORDS

Ant colony algorithm, ACS, particle swarm optimization algorithm, P-ACS