Industrial robot grabbing operations are an important application in industrial production. In traditional applications, the grasping operations performed by industrial robots are generally implemented through teaching or offline programming. The starting and ending postures of the grasping operation are preset. In this case, once the work object and the working environment change, the established grasping action will become invalid.

Visually cracking crawling problems

The vision system is used to obtain information about the workpiece and its surrounding environment on the production line, identify the target workpiece to be operated, and communicate the target data to the industrial robot based on Ethernet, guiding the industrial robot to complete operations such as grabbing and placing the workpiece. This greatly improves production efficiency and processing quality.

Install a camera and an end actuator at the end of the robot so that the product workpiece can fully appear in the camera's image.

System composition

The entire system includes vision system and control system:

1.Visual system

The camera is responsible for the collection of visual images and machine vision algorithms;

2. Control system

It consists of a robot controller and a robot control box, used to control the robot to grab products;

The system contains three units: image processing unit, communication unit and industrial robot unit.

working principle

The working principle and working process of the system are:

(1) When a product exists in the industrial robot grabbing working area, the camera takes pictures of the working area to obtain product image data.

(2) The robotic arm performs digital image processing on the collected images. The robotic arm has rich graphical visual algorithm programming and built-in rich 2D (positioning, scanning, detection, measurement), 3D (segmentation, positioning, detection) algorithms. By using the image recognition method, the characteristics of the target product are extracted, data recognition and calculation are performed, and the target product is visually positioned. And communicate the location information of the target product to the industrial robot.

(3) Use camera calibration technology to achieve calibration of the camera-robot coordinate system, and use inverse kinematics to solve the position error values of each joint of the robot.

The robot receives location information and implements intelligent grabbing

The robot control box is used to complete the robot's motion control and trajectory planning, control the end actuator of the industrial robot, adjust the robot's posture, grab the designated target product, and place it at the designated location.

In the modern automated production process, machine vision + robots have been widely used in various fields. The characteristic of machine vision systems is that they can improve the flexibility and automation of production. Robots are precisely a huge application platform for intelligence +. With the development and maturity of machine vision technology itself, it can be expected that the integration of machine vision and robots will be increasingly used in various industries in the future.