Abstract—The evolutionary computing based on Particle Swarm Optimization (PSO) technique has been proposed to obtain better performance for solving travelling salesman problems. Basically, the original PSO encounters a problem of convergence before tackling the best among local optimal solutions. To eliminate such problem, this paper presents an enhanced PSO algorithm called FOGS-PSO, which is a combination of PSO and Fast Opposite Gradient Search (FOGS) under benefits from the exploration ability of PSO and the ability to generate effective candidate solutions of FOGS. This algorithm is divided into two phases. Firstly, FOGS is applied to generate the best candidate solutions locating on the manifold of objective. Secondly, PSO is then applied to improve the searching result and speed. Travelling salesman problem was experimented as well as the objective function according to Hopfield-Tanks network. The proposed algorithm is compared with a variety of algorithms based on PSO techniques. The results of the test problems show that the algorithm performs well in terms of distance and number of generations.

Index Terms—Opposite gradient search, optimization, particle swarm optimization, travelling salesman problem.

Thirachit Saenphon is with Faculty of Information and Communication Technology, Silpakorn University, Nonthaburi 11120, Thailand.

Cite:Thirachit Saenphon, "Enhancing Particle Swarm Optimization Using Opposite Gradient Search for Travelling Salesman Problem," International Journal of Computer and Communication Engineering vol. 7, no. 4, pp. 167-177, 2018.