This is a vision-guided picking robot: it locates cans in real time as they move along the line and picks them up regardless of gaps, jams, or orientation changes, a problem that’s common on canning lines where cans rarely arrive perfectly spaced or aligned.
A fixed-position robot assumes every item arrives in the same spot, in the same orientation, every time, and on a canning line, that assumption breaks down quickly. The customer’s existing process relied on someone standing at the line correcting can position and orientation by hand, which is exactly the kind of task that’s tedious for a person and a poor use of their time over a full shift. Any automated solution also had to avoid denting or marking the cans, since handling damage at this stage shows up as a quality problem downstream.
A vision camera locates each can in real time, identifying its position and orientation on the line and feeding that information directly to the robot controller so the pick point is recalculated for every single can rather than assumed from a fixed grid. The end-effector is designed specifically to handle cylindrical product without denting or marking it, balancing a secure enough grip to pick reliably against a gentle enough contact to protect the can’s finish.
Because the picking decision is made dynamically rather than from a fixed program, the cell absorbs the kind of upstream variability (gaps, jams, orientation drift) that would otherwise need a person correcting it by hand. It’s a clear demonstration of what vision-guided picking is actually for: letting a robot handle the “look and adjust” judgement call a person would normally make, but at a pace and consistency that holds up for a full shift rather than degrading toward the end of it.