Adaptive feature extraction and mapping for mobile robots in unknown environment

Abstract

This paper proposes a new approach for feature and line extraction use in mobile robot mapping with high accuracy, efficient speed and low complexity to determine line boundaries. This algorithm called modified split and merge (MSAM) is robust against measurement noises and demonstrates satisfactory results on different surfaces. The method is based on the least square method to fit a line to a series of uncertain points. Different least square criterion is investigated to choose the best one for line extraction. A novel approach is proposed here to adopt threshold base split and merge method on different surfaces. The approach is applied to NAJI2 rescue robot for simultaneously localizing and mapping the observation of its performance. The results of this study proved that it has a good real-time capability to integrate the information of laser scanner into the navigation algorithm of the mobile robot; however, the extracted lines are suitable for object base mapping approaches.   Key words: Line extraction, mapping, modified split and merge, feature extraction, fitting.