ACO-PSO optimization for solving TSP problem with GPU acceleration

Abstract

In this paper, we present a novel approach named “ACO-PSO-TSP-GPU” to run PSO and ACO on Graphical Processing Units (GPUs) and applied to TSP (Parallel-PSO&ACO-A-TSP). Both algorithms are implemented on GPUs. Well-known benchmark problems for many heuristic and meta heuristic algorithms presented by Travelling Salesman Problem (TSP) are known as NP hard complex problems.TSP was investigated using classical approaches as well as intelligent techniques employing Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO). Parallel computing is well suited to the execution of nature and bio-inspired algorithms due to the rapidity of parallel implementation. Results show better performance optimization when using parallelism compared to results using sequential CPU implementation.