Geese-Inspired Hybrid Particle Swarm Optimization Algorithm for Traveling Salesman Problem

Abstract

A new improved algorithm called geese-inspired hybrid particle swarm optimization (geese-HPSO) was proposed based on the generalized PSO (GPSO) model and inspired by the characteristics of geese's flight. The new algorithm redesigned the updating operator for each particle as follows. For one thing, each particle intercrossed with the corresponding particle of the sorted population, which made the first particle acquire the best updating information so as to quicken the convergence speed greatly. For another thing, the foregoing crossed particle intercrossed with the particle which is ahead its corresponding one of the sorted population. That prevented all particles from being attracted by the global optimum only and flying to the same direction so as to strengthen the diversity of the particles and avoid falling into the local optimum. The simulation results of several benchmark TSP problems for both smaller-scale and larger-scale show that geese-HPSO algorithm not only has higher convergence precision and faster convergence speed but also is stronger and can search in the global scope effectively.