Simultaneous localization and mapping using swarm intelligence based methods

Abstract

The problem known as simultaneous localization and mapping is of fundamental importance both in its own right and because of its potential applications in the development of autonomous robots. While many solutions exist that are based on classical Newton-like optimization techniques regarding scan matching, relatively no work has been done with respect to the application of derivative free bio-inspired techniques to this particular area of robotics. That being said, we propose a novel approach to the scan-matching step within the Simultaneous Localization And Mapping (SLAM) problem based on the exploitation of swarm intelligence. For this purpose, we have chosen three swarm intelligence optimization methods to be the subjects of our research investigation, namely particle swarm optimization, artificial bee colony and the firefly algorithm. Aiming at reducing further the translational and rotational scan alignment errors, the proposed scan matching proceeds in two main steps, namely scan-to-scan and scan-to-map matching. Furthermore, we have made use of a pose graph based approach as a means of maintaining the consistency of our estimates across the mapping process. Systems designed to perform simultaneous localization and mapping using pose graphs are currently the state of the art and it is our belief that a robust scan-matching system is currently of the utmost importance to further the field. We tested the proposed solution in many scenarios. We conclude that the artificial bee colony strategy is efficient in a large range of circumstances. This affirmation is backed by the fact that in the best case scenarios, we have obtained good and some times better accuracy gains, regarding the translational and rotational estimates of the robot’s trajectory, by using the this meta-heuristic, when compared to state of the art SLAM systems. The firefly algorithm, while not as accurate as the artificial bee colony technique, was faster on 7 out of the 8 public domain datasets. The firefly algorithm consumed, on average less time spent per scan than by the artificial bee colony optimization technique. Particle swarm optimization has shown an inferior accuracy when compared to the artificial bee colony optimization technique and an intermediate processing time when compared to the other two optimization methods. So, firefly based meta-heuristic can be considered as the technique that provides a good trade-off between the high accuracy offered by the artificial bee colony and the high execution speed of the particle swarm optimization.