Abstract
This paper describes about the development of mobile robot and localization system. The mobile robot mechanical system is build based on 3 wheels type and it includes of 5 units of ultrasonic sensors and dead reckoning sensor at each drive wheel. The mobile robot used skid steered method for its mobility. The ultrasonic sensors reading are filtered using Extended Hessian Normal Form (EHNF) to represent a model plane. Based on UMBMark method, the encoder is tune to reduce the systematic error and non-systematic error. Finally, the localization of the mobile robot is based on Extended Kalman Filter (EKF) method to estimate the mobile robot position in a global map.
