Rao-blackwellized particle filters associating genetic algorithms for mobile robot SLAM

Bingrong Hong,Maohai Li,Songhao Piao

2008-01-01

Abstract:In this paper a solution framework based on Rao-Blackwellized Particle Filters (RBPF) associating Genetic Algorithms (GA) is proposed for recovering the full SLAM posterior of mobile robot using raw sensor measurements, i.e. without extracting landmarks using complex feature extraction methods and data association. The resultant Filter permits the uses of any arbitrary observation model. Since the proposed method represents the map state for each robot trajectory using a population of species instead of particles, which is much more memory efficient than FastSLAM through the application of dynamic data structures for representing the maps and the robot trajectories. This makes the proposed method very suitable for SLAM in large environments. Further, the adaptation of species population sizes and the alternative map representations makes the proposed method adaptable to varied environments and different sensors. Experimental results obtained in an indoor environment using a laser measurement system are presented to demonstrate the method's efficiency.