Tile climbing robot capable of adhesion and inspection by the same vibration source

Abstract

Traditional tile inspection robots perform wall adhesion and tapping inspection with separate driving sources, resulting in high load due to system complexity. In this study, we developed a wall climbing robot capable of wall adhesion and tile floating detection using the same vibration source. In real machine experiments, the robot performed tile floating inspection by deep learning while adhered to the tile. As a result, it achieved a detection accuracy of 82%.