A Geese PSO tuned fuzzy supervisor for EKF based solutions of simultaneous localization and mapping (SLAM) problems in mobile robots

The present paper shows how a recently proposed modified Particle Swarm Optimization (PSO) algorithm, called Geese PSO algorithm, can be utilized to tune a fuzzy supervisor for an adaptive Extended Kalman filter (EKF) based approach to solve simultaneous localization and mapping (SLAM) problems for mobile robots or vehicles. This type of fuzzy based adaptive EKF approach for SLAM problems has recently been shown to be an effective approach to improve performance in those situations where correct a priori knowledge of process and/or sensor/measurement uncertainty statistics i.e. Q and/or R respectively, is not available. The newly proposed system in this work is demonstrated to provide better estimation and map-building performance in comparison with those fuzzy supervisors for the adaptive EKF algorithm, where the free parameters of the fuzzy systems are tuned using basic PSO based algorithm. The utility of the proposed approach is aptly demonstrated by employing it for several benchmark environment situations with various numbers of waypoints and landmarks, where the Geese PSO algorithm could tune the fuzzy supervisor better than the basic PSO based algorithm.