Sunday, April 09, 2006

"All eyes on the line"

Industrial robots have evolved considerably since an invention by Joseph Engelberger and George Devol called “Unimate” was installed at a General Motors plant to extract moulded parts from die-casting machines in the late 1950s. Robots today have evolved to the point where they can be used for a variety of tasks with just a change of the tools and the programming. Their precision and accuracy have improved by leaps and bounds, as has the software that powers them, making these automation tools capable of handling tasks as varied as grasping and placing an auto windshield to selecting only perfectly formed cookies for packaging. Industrial robots are cropping up in almost every factory setting, being used for packaging, assembling, palletizing, dispensing etc.
Previous posts have given a deeper insight into the working of robotic arms. But, how do these robots gather accurate information to accomplish a task without error? In the industrial setting, parts and components must be locked into precisely fixed positions for "blind" robotics systems (consisting of the robot and a computer-driven controller) to function. To ensure such consistency and reliability in industrial processes would mean high-cost, custom-engineered and low-tolerance fixed manufacturing lines. Here, a development contributing to the smooth, flexible working of industrial robots is examined, that is the addition of a machine vision system to these robots. This system enables robots to adapt to changing conditions and variability in the production environment.
ABB U.S. partnered with Braintech Inc. in North Vancouver, British Columbia, to create a commercial out-of-the-box vision-enabled robotic system, called TrueView. The system consists of 1. An ABB robot (as shown in the figure on the right) 2. An ABB controller 3. Braintech's eVisionFactory (eVF) software 4. A computer running Windows 5. A standard CCD camera 6. An integrated lighting system to ensure the camera can capture clear images 7. An end-effector (i.e. a robot "hand") 8. Conveyors
TrueView, combining ABB robots and Braintech’s eVF software has helped liberate manufacturing industries from the limitations of fixed automation. The system functions as described below.
- A 3-D camera, which is integrated into the robot’s end-effector captures an image of an object as it moves along the conveyor, and transmits that image via an Ethernet network to the Windows-based PC.
- The eVF software on the PC analyzes the image to find identifiable features in the object.
- The software uses that information to calculate where the object sits in 3-D space (it defines the x, y, z position, and roll, pitch, and yaw angles) and transmits that coordinate data back to the robot, so the robot hand can intercept each part correctly for grasping or performing other processes, with accuracy to within one tenth of a millimetre!
The TrueView system can link up to five robots under one controller. It can see the differences among several different parts and perform multiple actions as circumstances might require.
In a nut-shell, the success of robotic arms (however flexible and evolutionary they may be) is primarily attributed to the competency and robustness of their computerized control systems. The arms themselves are mere puppets swaying to the tune of the controller...

1 comment:

Security said...

s0500296 Martin Wiig

There really seems to be great potential in machine vision for several aspects of robotics, but i think what will be first and foremost felt is the impact on automation industry, as described in this posting. Just think how tedious it must be to program robots to do excactly that at this at the carefully planned time table, just to have it fail at its task because the conveyour belt is a bit slow today. Instead the robot can see that the belt has an off day and say to itself (and its 4 other companions if it is linked to a system like trueview) "dear goodness, i appear to be able to have a good few microseconds off at last!". Just like an engineering student