Do I Need a Vision System for my Robot or Should I Fly Blind?

by Joe Palace, Senior Electrical Engineer

When a robot performs an application, a process, or a validation check on a part, it relies on that part being the same size and being in the exact location before it does its routine. But what if this part is not the same size or is not located where it is supposed to be? The end result is bad things happen, like:

  • A ‘crash’
  • Invalid data results
  • The process does not complete

Human intervention is needed to correct this problem. Hopefully, the problem can be corrected. If it cannot, then the system is down and productivity (and thousands of dollars) is lost.

But sometimes, the part that needs to be processed or checked is not always the same size or cannot be perfectly located in the same location every time. This is where a vision system comes into play.

When is using a Vision System a good idea?

Vision can:

  • Check the part size
  • Check the location of the part
  • Relay results back to the robot like part size or X, Y, Z coordinates of its location

The robot will adjust its path based on the results from the vision camera and then process the part for whatever application it is supposed to do. No production is jeopardized and no downtime is encountered.

There are also some cases where the robot might need to make a decision before executing its path. Say it has to decipher between two different parts because these parts get put into different dunnage. The decision is based off the successful read of a barcode stamped on each part. The robot, with vision capability, can execute this decision in a split second and correctly load the parts in the right dunnage every time, and all on its own. No human interaction is needed.

Another case for vision is the introduction of new parts, where they are not the same size as the old ones. It might take months before a new fixture is designed to house these parts. With vision, a new program is used to train the new part, which only takes a fraction of the time as it would for a new fixture design. Vision identifies the new part and then sends the correct job number to the PLC and or robot directly to process the correct part. A definite time savings here.

When is using a Vision System not a good idea? 

For robotic applications that process parts from a perfectly located fixture and placed on a pallet, that is firmly placed against a hard stop, using its palletizing algorithm within its own program, there really isn’t a need for a vision system because the parts are always where they should be and the end location where the parts go never changes.

Another example is a robot application in a very dirty environment where there is lots of dust. A vision system in a dirty environment is going to require maintenance from maintenance personnel to wipe the dust off the camera(s) so that the images are processed correctly before being relayed to the robot. This might not be cost effective for a company limited on resources.

Is Vision right for your Industrial Automation System?

Many determining factors are involved in making the right choice with your automation needs. With Vision Systems as an option, we’ve just touched on a few of them to consider. It’s not always about the cheapest solution but the right one!

It is important to consider how the following factors affect each option (vision system or blind) for your particular application:

  • More human interaction?
  • Increased downtime?
  • Decrease in production?
  • Less system reliability?

The cost of these factors add up and could be drastically more expensive than the investment of adding vision to your robotic system.

Would you like to discuss your project specifics?

Free Robot Assessment

Related categories: Blog Robotics

Terrance Brinkley's Bio

Michigan Director of Operations

With a natural affinity for control systems integration, Terrance Brinkley has been an asset to Patti Engineering since 2004 and now leads his team as the Director of Michigan Operations. A native of Pontiac, Michigan, Terrance graduated from Michigan State University with a Bachelor of Science degree in both Electrical Engineering and Computer Engineering.