Determination method of stable grasping parameters for irregular sheet sorting

Irregular sheet sorting is one of the most widely used mechanical production procedures, particularly in shipbuilding. This process typically involves handling thousands of sheets, each with its unique shape and orientation, and is often carried out manually. Though manual grasping is a common but inefficient and hazardous approach, the challenge of automatic irregular sheet sorting is the difficulty in grasping parameter determination. To address this challenge, we propose the determination method combing semantic segmentation with a graphic algorithm for ship sheet sorting. This method has implemented an automated sorting system capable of determining four-dimension grasping parameters. The semantic segmentation network is used to segment ship sheets in complex and cluttered scenes enabling the system to determine sheet position. The graphic algorithm determines the state of the grippers. The ship sheet automated sorting system comprises a robot device, a camera, and a matrix electromagnet tool for maximum suitability for all irregular sheets. Experiments and analyses were carried out, and results showed that the procedure is effective, and the determination method aids in the automation of shipbuilding.