The AIS-5000 is a commercially available massively parallel processor which was designed to operate in an industrial environment. It has fine-grained parallelism with up to 1024 processing elements arranged in a single-instruction multiple-data (SIMD) architecture. The processing elements are arranged in a one-dimensional chain that, for computer vision applications, can be as wide as the image itself. The overall architecture of the system is described. Various components of the system are discussed, including details of the processing elements, data I/O (input/output) pathways and parallel memory organization. A virtual two-dimensional model for programming image-based algorithms for the system is also presented. Performance benchmarks are given for certain simple and complex functions.>

Machine vision offers manufacturing industries a more cost-effective means of developing and implementing robotic applications. How using a vision sensor can increase the flexibility and usefulness of robots in specific manufacturing applicatons is described, provided that specific features of the robot-vision application development environment are present. These features include an object model representation that incorporates both statistical and relational information, a robot-vision interface that accommodates the flow of data and control between the vision sensor and the robot controller, and an integrated user progranunng environment that allows the development of object-oriented robotic applications. A system is described that incorporates these features and some requirements for future development of robot vision are given.