Flexible automation is a partial solution to America's industrial and economic recovery and survivability. At the core of this manufacturing technology is the robot whose own flexibility is enhanced through the use of computer vision to close the control loop on the arm and also give the robot environmental intelligence. The basic subdivisions of a computer vision system are sensing, segmentation, description, recognition, and interpretation. When the vision processor analyzes a scene it can give the robot controller symbolic object information and object coordinates. The robot controller transforms such coordinates into its own reference frame through the use of homogeneous transformation matrices. Vector formulations can be used for error signal information in visual servoing. These formulations will be in the form of differential translation and rotation transformation vectors. Variations of this technique and other approaches are being used in significant experimental systems at the University of Illinois, National Bureau of Standards, General Motors Research Laboratories, Hitachi of Japan, and others. Areas still requiring much research include imaging systems, processor architectures, dynamics formulations, algorithms and software, and artificial intelligence.
| Identifer | oai:union.ndltd.org:ucf.edu/oai:stars.library.ucf.edu:rtd-1689 |
| Date | 01 July 1983 |
| Creators | Janka, Randall S. |
| Publisher | STARS |
| Source Sets | University of Central Florida |
| Language | English |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Retrospective Theses and Dissertations |
| Rights | Public Domain |
Page generated in 0.0024 seconds