Mobile robot simultaneous localization and mapping in unstructured environments

This paper describes a method of mobile robot simultaneous localization and mapping (SLAM) based on laser range finder in unstructured environments. Considering drawbacks of the traditional iterative closest point (ICP) algorithm, matching laser range scans gets into local extrema when severe occlusions occur, the iterative dual closest point based on clustering (IDCP BoC) algorithm is proposed. Scan data are filtered and divided into clusters firstly, and then a procedure of reducing of data is conducted. The closest point (CP) rule is modified and dual closest point (DCP) rule is advanced when choosing the closest points. The reduced data is used for iterative computation before the error of two consecutive iterations' residual errors less than a preset threshold to speed up the algorithm, after that the data which is not reduced is used to guarantee the accuracy. Experimental results show that the method improves the accuracy of localization and mapping.

I. Introduction

The problems of simultaneous localization and mapping are of profound importance to the field of autonomous mobile robotics [1]. Self-localization of mobile robots is a procedure of estimating its position and orientation [2]. Map building refers to the process of estimating the position and shape of obstacles in the space where the robot is moving. The representations of environment map that can be divided into three categories: grid map, geometric map and topological map [3]. Geometric map is that robot collects the perception information of the surrounding environment, to extract abstract geometric characteristics, such as line segments or curves, and then describe the environment by using these geometric information, so the representation method is more compact and easier for mobile robot to estimate position and recognize target.