Performance Evaluation of the Grid-based FastSLAM in V-RFP Using MATLAB

This paper presents a Simultaneous Localization and Mapping (SLAM) application that is developed in V-REP robot simulation program by using Grid-Based FastSLAM method. In this work, the SLAM problem in an unknown indoor environment is solved with the Pioneer 3 DX mobile robot equipped with a laser range finder. Control scripts are developed in the Lua and FastSLAM scripts using MATLAB that is linked to the simulation platform by means of the Remote API feature of the V-REP. In order to evaluate of the performance of the FastSLAM, a number of tests have been conducted. Effects of the changing of particle numbers, changing in the error due to steep turn, changing of the number of sensor beam, cpu time for the iterations and the map error due to working time are analyzed in detail.