Monocular Vision-Based Parameter Estimation for Mobile Robotic Painting

Abstract

Interior wall painting is widely used in construction, installation repair, decoration, and other construction industry. By replacing human manual operation, robotic painting is developed to improve efficiency and reduce the cost. This paper proposes a painting parameters estimation scheme for robotic painting process. The parameter estimation system is built on a single camera, a laser level, and four laser distance sensors. The camera's intrinsic parameters are obtained by a system calibration method. The relative position between the robot and the wall is obtained through the single-image measurement. The wall plane equation and the position of paint-start point are calculated from the measurement system. The experiments are conducted in actual interior wall painting task for a construction robot. Compared with other methods and systems, the proposed scheme for painting robot demonstrates superior accuracy performance and is effective, simple, and easy implementation.