Escaping Local Minima Using Repulsive Particles in FastSLAM for Space Rover

Daiki Koike and Kenji Uchiyama
Nihon University, Chiba, Japan
Abstract—This paper describes the efficient strategy for guidance and control of a space rover using estimated position and a map acquired by using FastSLAM (Simultaneous Localization and Mapping) algorithm using a particle filter. A guidance law based on potential function method is also designed to cope with topographical change on a planet. Using the method, local minima problem in the potential field are frequently occurred when designing repulsive potential functions so as to avoid obstacles. To overcome the problem, we propose the new method that generates repulsive potential function at particles in FastSLAM for an exploration rover in unknown environment. Numerical results are performed to verify the validity of the proposed method for solving local minimum problem. Adaptive temperature parallel simulated annealing (ATPSA) method that is usually used to escape local minima is also applied to the system for comparison.

Index Terms—component; FastSLAM; potential function method; repulsive particles; particle filter; space rover

Cite: Daiki Koike and Kenji Uchiyama, "Escaping Local Minima Using Repulsive Particles in FastSLAM for Space Rover," International Journal of Mechanical Engineering and Robotics Research, Vol. 6, No. 6, pp. 506-511, November 2017. DOI: 10.18178/ijmerr.6.6.506-511
Copyright © 2016-2017 International Journal of Mechanical Engineering and Robotics Research, All Rights Reserved
E-mail: ijmerr@ejournal.net