Hi,
For a school project we are looking to integrate Robotic wrist camera on our UR3 robot and we are not sure if our lab idea is feasible.

The main idea is: a conveyor feed the robot with a squared part with a printed form on it (square, triangle, star, etc.). Parts are fed in a random order. We want to be able to detect the type of bloc it is and sort it in different piles.

If we use 3 camera nodes, one for each part (A, B, C), to detect part C the program has to take a picture to search for part A (2-3 Sec.) and part A is not found. So, the camera takes a new picture to search for part B (2-3 Sec. again) and finally take a picture of part C and found it. The process time is quite long.

Is it possible to speed up the process?

Thank you for your help.

Tomy