Robotics is becoming increasingly attractive even for simple tasks. However, obstacles are the programming effort and the expertise needed to enable the robot to ‚see‘ i.e. to detect the parts. Visor Robotic is a compact and easy to use 2D vision sensor for robot guidance. It consistently uses 3D poses, so that no complex transformations from 2D to 3D have to be programmed by the user. With the function „3D result offset“ the relation between the recognized 2D image position and the desired 3D gripping position can be taught by moving to this position once. The new detector „Contour 3D“ even allows a real and precise 3D localization by perspective changes in the image. Four calibration methods allow the simple solution of complex tasks, whether the camera is fixed or mounted on the tool. It is sufficient to perform a single calibration, no matter at which position the robot is working. Such specializations increase the availability of the machines and thus contribute to cost savings.
This solution was nominated for the Automation Award 2020 in the TECHNOLOGY category.