Keypoints based laser scan matching — A robust approach

Abstract

This paper aims at development of a robust keypoints based scan matching (KSM) methodology for 2D laser data applied to mobile robot navigation. In this method feature points are first transformed to grid points, and then represented in the form of image. Keypoints are extracted using Harris corner detection method and finally matching is done by RANSAC method. Real world experiments have been carried out to evaluate the utility and robustness of the proposed method. A comparison has been drawn with scan matching technique with ICP method. The proposed method might be used in a robust manner for mobile robot navigation with moderate computational complexity.