Home /Research /Distributed manipulation environment—(a transputer-based distributed robotic system)
MANIPULATION

Distributed manipulation environment—(a transputer-based distributed robotic system)

Fazel Naghdy, Paul Strickland

Year
1989
Citations
5

Abstract

Abstract In this paper a new approach for design and development of robotics system referred to as Distributed Manipulation Environment (DME) has been proposed. The major drawbacks in the employment of conventional robot arms in complex tasks have been addressed and it has been shown that DME may overcome the deadlock facing the engineers and industrialists in the field of robotics. During the course of the paper, the technique is analysed and its advantages, compared to conventional robotic systems, are discussed. DME is shown to provide a more general view on manufacturing systems and can integrate robotics systems with other dedicated and semi-dedicated machines conventionally used in production. An alternative approach for parallel/distributed processing of DME based on Occam/transputer has been proposed. It has been shown that the task of software design and development based on this architecture is simpler and more systematic than conventional methods of distributed processing. In this approach the software controlling a manufacturing cell or a production line can be programmed entirely in Occam, tested in pseudo-parallel processes on one transputer and mapped out on to the transputers driving each Manipulation Module. The work conducted in developing two transputer-based boards for DME named as ‘MM Controller Node’ and ‘Processor Node’ is also reported, and the progress to date for implementing the DME concept for an industrial robot as a super-node is given.

Keywords

TransputeroccamRoboticsComputer scienceArtificial intelligenceNode (physics)SoftwareDistributed computingRobotParallel processing

Related papers

Browse all MANIPULATION papers