Robust laser positioning in a mobile robot machine vision system

Abstract

Laser scanning devices are widely used in Machine Vision Systems (MVS) of an autonomous mobile robot for solving SLAM problems. One of the concerns with MVS operation is the ability to detect relatively small obstacles. This requires scanning a limited sector within the field of view or even focusing on a specific point of space. The accuracy of the laser beam positioning is hampered by various kinds of uncertainties both due to the model simplifying and using inaccurate values of its parameters, as well as lacking information about perturbations. Goal. This paper presents the improvement of the MVS, described in previous works of the authors, by robust control of the DC motor, which represents the Positioning Laser drive. Methodology. For this purpose, a DC motor model is built, taking into account the parametric uncertainty. A robust digital PD controller for laser positioning is designed, and a comparative evaluation of the robust properties of the obtained control system with a classical one is carried out. The PWM signal formation by the microcontroller and processes in the H-bridge are also taken into account. Results. The obtained digital controller meets the transient process and accuracy requirements and combines the simplicity of a classic controller with a weak sensitivity to the parametric uncertainties of the drive model. Originality. The originality of the paper is in its focus on the MVS of the autonomous mobile robot developed by the authors. Practical value. The implementation of the MVS with the proposed controller will increase the reliability of obstacles detection within a robot field of view and the accuracy of environment mapping.