Mobile Robot Localization Based on Extended Kalman Filter

The mobile robot localization methodologies in common use at present have been introduced. A localization algorithm based on extended Kalman filter (EKF) has been proposed on the basis of environment feature extraction and map building, which can reduce the error in the calculation of the robot's position and orientation. The method is that the mobile robot analyses and fuses the messages in surroundings from multiple sensors by EKF theory, which enables the robot to identify the surrounding objects clearly and guide itself successfully. The simulation and experimental results show that the proposed localization method is effective

I. Introduction

In order to navigate safely and reliably, an autonomous mobile robot must be able to find its position simultaneously within its environment. To date, there have been many localization methods with respect to the work condition complexity, the category and number of the mounted sensors. All these methods of localization can be divided into two main categories: the relative and the absolute [1], [2]. Relative (local) localization consists of evaluating the position and the orientation through integration of information provided by diverse sensors. The integration is started from the initial position and is continuously updated in time. Absolute (global) localization is the technique which permits the robot to find its way directly in the domain of evolution of the mobile system. These methods usually rely on navigation beacons, active or passive landmarks, map matching or satellite-based signals like Global Positioning System (GPS).