Why do robots need machine vision systems?

1. The application of robots without machine vision has limitations. The
position must be fixed
on the production line assisted by robots. The movement position of the robot is preset according to the characteristics of the production product, that is, the movement is performed according to the preset movement trajectory. Therefore, certain fixtures are needed to ensure the fixed position of the product. If the product specifications are diverse, a large number of special fixtures are needed to fix the product position. The cost is high and the workload of replacing the fixture is huge.
The contradiction between accuracy and reliability.
The higher the accuracy requirements for robot assembly, the higher the accuracy of the fixtures required. The higher the accuracy of the jig, the more difficult it is to assemble the robot body accurately and reliably every time.
Disadvantages of contact positioning:
Due to the delicate surface of some products, they cannot be supported by mechanical clamps to avoid damaging the surface. Some products cannot be grasped reliably because they are made of flexible materials.
2. The machine vision system helps position modification for robot applications . The vision system is used to inform the robot of the position of the product and provide the positioning information for grabbing. When replacing the product, it only needs to be replaced. Product testing documents are sufficient. Save a lot of machinery costs and time required to replace fixtures. To ensure accuracy through multiple positioning, low-precision fixtures or rough positioning vision systems can be used to enable robots to complete the work of grabbing products. After grasping, the characteristics of the object are accurately captured through the vision system to achieve high-precision positioning, so that the robot can further correct the position after grasping and achieve precision assembly. Non-contact measurement  machine vision systems use optical measurement methods, which will not damage the surface of the object and will not prevent it from being measured because the object is flexible.

 3. The limitations of a single machine vision system in production cannot be fully observed . From the principle of imaging, a camera can only capture image information on one plane. For complex objects that need to detect multiple faces, many cameras often need to work together. If the product has many specifications, cameras with different specifications need to be adjusted to different positions for detection. This makes the entire detection system extremely complex. The contradiction between accuracy and field of view is limited by the resolution of the camera's photosensitive chip. The larger the field of view, the lower the resolution accuracy. In this case, high-precision detection cannot be achieved for large objects. Possibility of multi-posture By installing the vision system on the joints of the robot, the robot can be used to adjust the position of the camera or light source to meet the detection requirements of various postures, thus realizing the detection application of multi-specification and complex products. Follow-up detection can achieve high-precision detection of large objects in a small field of view by using a robot to carry a camera to traverse various detection parts of a large detection object or follow the movement trajectory of the robot for real-time detection.