Browsing through Design News, I came across mention of an interesting vision sensor from Omron. The FQ-M is designed to identify shapes on a conveyor. An encoder input allows it to track position and let a robot know where to find each item.
The literature shows this working with a robot similar to the ABB Flexpicker, though I imagine it would work with other robots and handling devices too.
As the Sysmac Studio software must be purchased separately, I suspect it all works out quite expensive for a single camera. Jump up to multiple lines though and the economics probably gets a whole lot more attractive.
Posts
4 comments:
What's the idea? Robot needs the encoder input anyways if picking moving objects.
Agreed, although sometimes the belt will slip or the object will move on the belt.
Any Omron sales people to clarify? :)
Using the encoder the camera can track products on images and distinguish the products on the images. i.e. when two images are shot the vision system checks the encoder count to calculate if the products on the second image are new or already shown on the first image. This reduces the amount of work you have to do when programming such a system.
Regards,
Harro Bosch PMM Automation Omron
Post a Comment