Positioning of a mobile robot based on odometry and a new ultrasonic LPS

• Published in: International journal of control, automation, and systems Vol. 11; no. 2; pp. 333 - 345
• Main Authors: Cho, Bong-Su, Seo, Woo-Jin, Moon, Woo-sung, Baek, Kwang-Ryul
• Format: Journal Article
• Language: English
• Published: Heidelberg Institute of Control, Robotics and Systems and The Korean Institute of Electrical Engineers 01-04-2013; Springer Nature B.V; 제어·로봇·시스템학회
• Subjects: Analysis; Code Division Multiple Access; Compasses; Control; Control systems; Correlation analysis; Encoders; Engineering; Errors; Field programmable gate arrays; Interpolation; Kalman filters; Mathematical models; Mechatronics; Receivers; Regular Papers; Robotics; Robots; Sensors; Studies; Transmitters; 제어계측공학; mobile robot; odometry; ultrasonic LPS; Kalman filter; positioning
• ISSN: 1598-6446; 2005-4092
• DOI: 10.1007/s12555-012-0045-x

Odometry is a method that calculates the position and heading angle of a mobile robot using encoders attached to the wheels of the robot. Errors in the position and heading angle in odometry continuously increase as the operating time and moving distance increase. The solution to overcome these accumulated errors is to periodically compensate with the external absolute position information. An ultrasonic local positioning system (LPS) consists of multiple ultrasonic transmitters located in the environment and an ultrasonic receiver. In this study, ultrasonic transmitters are in a line at one side, and four transmitters are grouped for a coverage area. In order to measure the time of flights (TOFs) for an ultrasonic signal, the receiver predicted the transmitted time from each transmitter using a hyperbolic model. Four transmitters emit ultrasonic signals sequentially, and then the receiver calculates the position using the present measured distance and the pre-measured distance. In order to extend the distance that is measured, the receiver collects the ultrasonic signal and executes cross correlation with a sinusoidal signal. The measured distance data of the previous step causes the position error. This error is compensated for by the predicted distance data using a bilinear interpolation method. An extended Kalman filter is designed to combine odometry, a compass sensor, and an ultrasonic LPS. The proposed system provides reliable and accurate position and heading information, regardless of the operating time and moving distance.