Discuss Proven technologies that addresses

Transcription

1 Robotics and Machine Vision for assembly -Auto Teach, Vision guidance, Color & 3D Mar Agenda Discuss Proven technologies that addresses o Quick Tool Bring up o Using Non-touch Vision based Auto Calibration/Auto Teach o Cycle time Reduction and increasing throughput o Using Vision on-the fly technology o High performance fast robot mechanism o Increase Precision of assembly o Using Vision Guidance with highly repeatable robot o Integrated Vision and Motion controller o 3D part picking o Using 2-camera 3D vision technology 1

2 Why Should I Use Machine Vision in Assembly? Vision guided applications don t require tight fixtures or trays for parts entering the system Pre-orientation of parts is not required Quality improves because parts can be inspected prior to and/or during assembly instead of after assembly Changes in part geometry only require a simple re-teaching of vision instead of a change in tooling Eliminate Assembly tolerance stack-up Eliminates Human error Accuracy of robot mechanism become close to its repeatability Flexibility Flexibility Flexibility Flexibility and cost containment! The cost of complexity Technical risk and cost go hand-in in-hand Synchronization Tolerance Communications User interface Calibration Risk increases as a multiple of complexity 2

3 Vision Calibration Three primary pieces of information determined by Calibration The millimeter-to-pixel ratio Correction factors for perspective distortion The camera-to-robot transformation Vision Calibration compensates for: Non-square pixels Camera not perpendicular to work surface Lens distortion The accuracy of a Vision Guidance System can Only be as Good as the Initial Calibration Quick Tool Bring up- Auto Calibration /Auto teach Vision based NON-Touch Tool calibration o Upward looking camera for Tool Tip calibration 3

4 Quick Tool Bring up- Auto Calibration /Auto teach (Contd.) Vision based NON-Touch Robot to camera calibration o Downward looking camera calibration Quick Tool Bring up- Auto Calibration /Auto teach (Contd.) Final location of part automatically found. No operator teaching required. Part location = Calibration Offsets(Down camera calibration) + Vision loc(picture from Camera) + Gripper offsets(upward camera) 4

5 Inspect and locate parts simultaneously Media inspect and Locate All good Missing & Cross slotted Vision on-the-fly "Vision on the fly" is a feature that allows the vision system to take a picture while the robot is moving the part through the camera's field-ofview. Once the picture is acquired, the robot proceeds to a tool refinement location, stops at that location while a tool refinement is performed, and then proceeds to the place (drop) location. 5

6 Vision on-the-fly Process Part Pick up from Cassette/Feeder Start Moving towards the destination Vision task waits till part is in the field of view of camera o Acquire image using upward looking camera and latch encoder simultaneously o Calculate robot tool offsets using vision image Approach final location o Use vision offsets and accurately position robot over the placement location Place part Color Vision Support for color cameras Color edge detection for all edge tools Color edge detection for Locator Color discrimination option for Locator to locate objects in color Color Matching tool to perform advanced color inspection Color calibration to ensure accurate color identification and rendering under various lighting conditions 6

7 3D Is the next big step in Vision Guidance Key features: Part location with 6 degrees of freedom Handling of parts placed at random Handling of overlapping parts Identification of pick able parts Collision avoidance 7