Workpiece grabbing method and device, computer equipment and storage medium

Abstract

The invention discloses a workpiece grabbing method. The method comprises the following steps that S1, an original point cloud of a workpiece is obtained; S2, the original point cloud of the workpiece is identified so as to obtain multiple pieces of workpiece information; and S3, a first grabbing planning module polls the workpiece information to conduct grabbing planning judgment, if all the workpiece information corresponding to all grabbing postures does not pass grabbing planning, operation turns into the step S4, and a second grabbing planning module polls the workpiece information to conduct grabbing disturbance judgment. By means of the method, the grabbing success rate can be effectively improved, and workpieces can be grabbed in a completely-disordered state.

Description

technical field [0001] The invention relates to a system and a control method for grabbing parts from a material frame, in particular to a workpiece grabbing method, device, computer equipment and storage medium. Background technique [0002] In the field of industrial manufacturing, the traditional loading and unloading system is realized based on automated tooling or manual methods. The workpiece is heavy, the manual operation is labor-intensive, and it is difficult to recruit corresponding workers. With the promotion of industrial-grade 3D cameras and the rise of 3D vision algorithms, the disorderly sorting and loading system based on 3D vision is expected to realize intelligence in the past manual or automated methods, and realize the industrial upgrading of automatic loading and unloading systems. [0003] The 3D visual guidance application grabs parts, grabs from the material frame stack, grabs from the accumulation chain, and grabs parts from the EHB conveyor system, ...

AI Technical Summary

Problems solved by technology

Second, due to the random placement of workpieces, it is easy to cause a large number of workpieces in the material frame to have overlapping, occlusion, shadows, and complex backgrounds. The complete point cloud data of such workpieces cannot be obtained through the camera, and the related workpieces cannot be identified, resulting in this. class artifact grabbing

Third, in the current disordered sorting, the gripper at the end of the robot is designed as a long straight rod, and the gripping strategy is to pick up the end perpendicular to the surface of the workpiece, such as figure 1 As shown, the pose of the robot is directly related to the straight rod angle θ (the angle between the straight rod and the z′-axis direction of the material frame). Insufficient development, interference with the material frame, etc.

like figure 2 As shown, when the workpiece is tilted by 30°, the angle θ of the straight rod has reached 30°. If the tilt angle of the workpiece increases, the angle θ of the straight rod will also increase, which will cause problems such as insufficient arm extension of the robot and easy interference with the material frame.

This grasping strategy and hand grasping design can only grasp workpieces with an inclination angle within ±30°

Images