Adding Flexibility, Lowering Costs With Robotic Vision Systems
Add a vision system to a robot and you give it eyes, the robot can find objects in its working envelope — reducing the need for complex and expensive fixtures. This increases the flexibility of robotic automation, adapting to variation in part size, shape and location — ultimately reducing cell complexity. By adding a vision system robot palletizing and de-palletizing becomes practical, components can be assembled, dies unloaded, and parts lifted off racks and out of bins. It's even possible to track and pick parts in motion on a conveyor.
How Does Robotic Visioning work?
Robotic vision systems consist of one or more cameras, special-purpose lighting, software, and a robot or robots. The camera takes a picture of the working area or object the robot will grip and software searches the image for features that let it determine position and orientation. This information is sent to the robot controller and the programmed positions are updated. Depending on the application, the camera might be mounted on the robot or could be in a fixed position within the cell. Calibration is usually needed to relate the vision system coordinate space to the robot. Robot companies like FANUC offer robotic vision systems with vision software tightly integrated with the robot controller, so simplifying programming and use.
Many applications only need X- and Y-axis information plus rotation — this is easily extracted from an image. Sometimes though, it's important to have height information too, for example, when unloading a pallet, especially if the cartons or bags differ in size, or when picking parts from a bin. There are several ways to get height information, such as using stereo cameras. However, in many cases, laser triangulation or part size are the best options. In laser triangulation, height is derived from the line position projected onto the target surface and viewed from an angle by a camera. One limitation though is that either the part or the laser line must move. Alternatively, when the part size is known, camera distance can be determined by the part’s appearance in size. This technique is common in vision guided de-palletizing.
Conveyor tracking is often useful in packaging applications. Product moves into the robot cell on a conveyor without stopping. A vision system determines accurate belt position and the robot controller adds speed information from the belt encoder — letting the vision system robot track and pick while the product is in motion. This eliminates any need to stop the belt or use expensive fixturing. Rather than presenting every piece in precisely the same location and orientation for the robot to pick up, a vision system simplifies cell design and can lower costs, making robotic automation even easier to justify.
To discuss known automation opportunities or discover new ones: