关键词: 海湾扇贝, Delta机器人, 定向上料, 耳部朝向, 响应面, 末端执行器, 视觉系统

Abstract:  At present, the infrared heating of bay scallops usually uses manual feeding before shelling and shell extraction, and the feeding accuracy and production efficiency are low. The irregular surface of scallops is the difficulty in realizing automatic feeding. This paper designs the automatic directional feeding system for bay scallops. Firstly, when the scallops transported by the conveyor belt arrive at the identification area, the visual system uses the RANSAC algorithm to fit the straight line of the scallop ear edge to determine the ear orientation of the bay scallop. Secondly, the conversion of image information coordinates to robot coordinates is realized through the data fed back by the conveyor belt and the encoder, and the actual grasping position is determined according to the trajectory information of the robot end. Finally, according to the irregular curved surface and multiple radial ribs of the scallop shell, the grasping method of the Delta robot end effector is determined. By using Box-Behnken to conduct multifactor interactive test analysis, it is obtained that when the diameter of the suction cup is 20mm, the grasping point is 023 times the height of the shell from the center line of the scallop to the back edge, and the input air pressure of the vacuum generator is 0.6MPa, the success rate of grasping Highest. Experimental verification is carried out according to the optimization results, and the results show that the grasping success rate is 95.80%, the theoretical value and the actual result are in good agreement, and the average grasping time is 1.6s. The success rate of ear orientation recognition reaches 95.33%, and the total success rate of feeding system is 91.67%. The research results can provide a reference for the field of scallop automatic processing.

Key words: bay scallop, Delta robot, directional feeding, ear orientation, response surface, end effector, visual system