Sensor guided robot path generation for repair of buoyancy module

Abstract

In the marine & offshore industry, the repair of buoyancy modules involves removal of defective regions which is commonly conducted manually using hand-held abrasive tools. This manual process is one of the key bottlenecks of the production process and the removed material can be hazardous to operators' health. To lower manpower costs, and together improve efficiency while minimizing the disruption on current work flow, a sensor guided robot material removal system is proposed to replace manual operation. The challenge in this automation solution is to develop an automated material removal solution for this high-mix-low-volume repair problem with minimum human input required and minimum impact on the current process. To address the challenges, a hand-toeye vision system is proposed to identify the shape, depth, size and location of the defective regions in order to improve the flexibility and robustness of the robot material removal system. Experiments showed that the proposed method could detect the defective regions robustly and remove them effectively.