A Hybrid PSO-ACO Algorithm for Precise Localization and Geometric Error Reduction in Industrial Robots

Abstract

The proposed hybrid optimization algorithm integrates PSO with ACO for the improvement of a number of pitfalls within PSO methods traditionally considered and/or applied to industrial robots. Particle Swarm Optimization may frequently suffer from local optima and inaccuracies in identifying the geometric parameters, necessary for applications requiring high-accuracy performances. The proposed approach integrates pheromone-based learning of ACO with the D-H method of developing an error model; hence, the global search effectiveness together with the convergence accuracy is further improved. Comparison studies of the hybrid PSO-ACO algorithm show higher precision and effectiveness in the optimisation of geometric error parameters compared to the traditional methods. This is a remarkable reduction of localisation errors, thus yielding accuracy and reliability in industrial robotic systems, as the results show. This approach improves performance in those applications that demand high geometric calibration by reducing the geometric error. The paper provides an overview of input for developing robotics and automation, giving importance to precision in industrial engineering. The proposed hybrid methodology is a good way to enhance the working accuracy and effectiveness of industrial robots and shall enable their wide application to complex tasks that require a high degree of accuracy