A dead reckoning sensor system and a tracking algorithm for mobile robot

Abstract

This paper presents a dead reckoning sensor system and a tracking algorithm for mobile robot localization which increases an accuracy of the estimating position at the uneven surface. A dead reckoning sensor system consists of a small, low-cost MEMS IMU and optical navigation sensor that used in laser mice. A tracking algorithm consists of multi-rate extended Kalman filter to fuse redundant and complementary data from IMU and optical navigation sensor. Also method for estimating angular velocity to estimate more accurate yaw bias is proposed. All the experiments are performed on the radio controlled car mounted with sensor system.