SVM Based SLAM Algorithm for Autonomous Mobile Robots

Abstract

Support vector machine (SVM) is a classification algorithm with some advantages over other machine learning methods, which provides an efficient tool to select new features from sensor observations. This paper presents a SVM based simultaneous localization and mapping (SLAM) algorithm that enables autonomous mobile robots to operate in a dynamic or unstructured environment. The observation models and the SVM based visual feature processing algorithm are designed. SVM is adopted in several steps of observation in this paper in order to achieve fast processing and accurate localization. The simulation results are given to show its feasibility and good performance.