GPU PSO and ACO applied to TSP for vehicle security tracking

Abstract

The Travelling Salesman Problem (TSP) is a well-known benchmark problem for many meta-heuristic algorithms, including security traffic optimization problems. TSP is known as NP hard complex. It was investigated using classical approaches as well as intelligent techniques using Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO) and other meta-heuristics. The Graphic Processing Units (GPU) is well suited to the execution of nature and bio-inspired algorithms due to the rapidity of parallel implementation of GPUs. In this paper, we present a novel parallel approach to run PSO and ACO on GPUs and applied to TSP (GPU-PSO&ACO-A-TSP) for security tracking vehicles in road traffic. Both algorithms are implemented on the GPUs. Results show better performance optimization when using GPUs compared to results using sequential CPU implementation.