Extrinsic calibration of complex machine vision system for mobile robot

Abstract

Autonomous mobile robots are essential in industry, logistics, and service applications, such as infrastructure inspection, where navigation in dark and narrow environments is one of the most challenging tasks. This project presents the design and development of a patented artificial vision system for a mobile robot, which combines a high-precision laser scanner and a stereo-vision system. The objective is to enhance the robot’s surface scanning capabilities in complex environments. The proposed methodology allows to determine the relative position between systems to integrate data of the laser scanner with stereo-vision cameras and inertial sensors, providing robust and accurate environmental perception that enables the characterization of ventilation ducts surface. Experimental results show a mean correspondence error of $7.4515\mathrm{mm}$ in absolute terms for the combined data.