A localization system of a mobile robot by fusing dead-reckoning and ultrasonic measurements

Abstract

This paper develops a novel system hardware structure and systematic digital signal processing algorithms for self-localization of an autonomous mobile robot by fusing dead-reckoning and ultrasonic measurements. The multisensorial dead-reckoning (DR) subsystem is established based on the optimal filtering by first fusing heading readings from a magnetic compass, a rate-gyroscope and two encoders mounted on the robot wheels, thereby computing the dead-reckoned location estimate. The novel ultrasonic localization subsystem consists of one ultrasonic transmitter and one radio-frequency (RF) controlled switch mounted on the known location fixed to an inertial frame of reference, four ultrasonic receivers and one RF controlled switch installed on the mobile robot. Four ultrasonic Time-of-Flight (TOF) measurements together with the dead-reckoned location information are fused to update vehicle's position by utilizing the extended Kalman filtering (EKF) algorithm. The proposed algorithms are implemented by using a host PC 586 computer and standard C++ programming techniques. A built system prototype together with its experimental results is used to confirm that the system not only retains its strengths of high accuracy and magnetic interferencing immunity, but also provides a simple and practical structure of use and installation calibration.