A Genetic Algorithm Based on Inter-Group Competition and Collaboration for the Traveling Salesman Problem
Qiang Zheng · Computer Engineering and Science · 2008
The classic traveling salesman problem is one of the NP-hard combinatorial optimization problems,and a fast and effective solution to TSP has the important theoretical and practical significance.Enlighted by the mutual contacts between the nature species groups,the paper proposes the inter-group competition and collaboration of genetic algorithms to solve the traveling salesman problem.In the iterative process,we only choose a competitive group to execute the evolutionary process each time.Meanwhile in order to maintain the balance between all the groups,the populations have a cyclical exchange,so as to promote good gene patterns to come into being in the process of evolution,spread in various populations and enhance the overall speed of evolution.This algorithm is not only effective for maintaining the diversity of the population,and will increase the speed of convergence.Simulation results show that the algorithm is feasible and effective.