Robot Mapping Using k-means Clustering Of Laser Range Sensor Data

Abstract

In this paper we discuss the map building technique for mobile robot using k-means clustering using laser ranger sensor. Clustering is an efficient technique to group together data set to obtain accurate maps for autonomous mobile robots. We discuss the k-means clustering algorithm in detail with experimental results and how this method can be used to obtain straight line maps for indoor environments. We discuss our results with different sizes of clusters and for data set with noise. Results confirm that k-means clustering can be used to obtain straight line maps for mobile robots.