点群レジストレーションによる検査箇所推定に基づくロボットアームでの膜厚検査の自動化

抄録

This paper describes the fundamental experiment for automating the inspection of a paint coating thickness of spline plates on bridge with a robot arm. The inspection point is estimated using the registration between 3D point cloud data obtained from the actual model and that generated from the CAD model. In the experiment, a robot is set on unstable scaffolds to imitate the inspection environment. The experimental results show that the estimation of the inspection point effectively works, and the inspection can be achieved, even when the robot and the bridge model are not perfectly facing each other. Besides, the results clarified the issues that the robot swung back and forth during the operation due to the unstable scaffolds, and the modification of the trajectory or attitude of the robot arm is required.