If you are considering installing robotic visual guidance systems on your industrial robots, you are likely curious about how your robot will recognize and locate objects.
In other words, how will you teach your robot what an object is (recognition) and where it is (localization)? You will do this quickly and easily with CortexRecognition®, the revolutionary software behind our breakthrough visual guidance system.
CortexRecognition®
CortexRecognition® is the software “brain” that allows an ordinary camera to be able to recognize a part. It acts as the interface between what the camera is seeing and what it needs the robot to do.
CortexRecognition® examines one image of your part using a 2D camera and without the need for intensive engineering. The software is capable of recognizing up to six degrees of freedom (X, Y, Z, Rx, Ry, Rz). It uses one eye, one image, and requires no structured light to calculate depth, rotation and location quickly and accurately. CortexRecognition® is compatible with every type of industrial robot (TCP or Serial) and works with complex 3D parts.
How You Teach Your Robot a New Part
Teaching your robot to recognize new parts requires no CAD courses and zero complex procedures. Simply present the part to the camera. Within a few minutes, CortexRecognition® recognizes and remembers the part or object, regardless of its orientation. This process takes five quick steps using simple point-and-click object teaching.
Step 1. Ready: Position the object in front of the camera. The object should be placed on a flat surface with a plain background – or, if it is a larger piece, it should be stood up in an area with minimal background distractions. This helps the camera to be able to focus on the part without having any disruptive elements, such as shadow and glare, that might get confused later in the process.
Step 2. Aim: Point the camera on the robot at the object, being sure to include all of the object within the camera frame. Since the system needs to understand where the part’s edges are, the entire piece should be visible within the frame. The camera can be zoomed in or out as needed to ensure the entire part is visible.
Step 3. Teach: On the screen of your computer or tablet with the CortexRecognition® software, press the “Teach” button.
Step 4. Erase: Once the camera has acquired its image, the software will identify individual data points that it recognized. Any data points that are outside of the part should be erased. The data points indicate to the software what it should focus on. The more detail that is present in the background, the more you will need to erase, which is why it is helpful to place the part on a solid backdrop.
Step 5. Guide: Once you have eliminated anything that is not a part of the piece you are working on, you are now ready to measure the distance to the object and start guiding the robot.
With the system programmed, it can quickly identify the point on your part that it needs – whether it is looking for a specific surface to attach to, or a specific point of reference on the part, it is able to identify it quickly and accurately – down to 0.5 millimeters.
It’s Really That Simple
CortexRecognition® software requires just one camera, one piece of software and no engineering. You simply point, click, and clean up. CortexRecognition® redefines robotic visual guidance by giving blind robots true sight and hand-eye coordination. Paired with our hardware system Robeye®, it quickly upgrades any automated line.
Contact us today to learn more about just how easy CortexRecognition® is to use and how quickly it can enhance your robots with true robotic guidance.
For more information, fill out the form below or call us to request a demo, have your questions answered, or speak to a representative.