Date of this Version
This document proposes the development of the remote site workcell for teleoperation with significant communication delays (on the order of one to 20 seconds). In these situations, direct teleoperation becomes difficult to impossible due to the delays in visual and force feedback. Teleprogramming has been developed in order to overcome this problem. In teleprogramming, the human operator interacts in real time with a graphical model of the remote site, which provides for real time visual and force feedback. The master arm and the manipulator/model interactions, given predefined criteria of what types of motions are to be expected. These commands are then sent via a communication link, which may delay the signals, to the remote site. Based upon a remote world model, predefined and possibly refined as more information is obtained, the slave carries out commanded operations in the remote world and decides whether each step has been executed correctly.
The remote site receives commands sent via the delayed communication link. These commands must be parsed and translated into the local robot control language, which includes insertion of dynamic parameters that are not generated by the master system. The commands are then executed by the hybrid position/force controller, and the resulting motions monitored for errors.
This proposal addresses the following remote site issues: low level manipulator control using an instrumented compliant wrist for sensory feedback, higher level command execution implementing dynamic parameters, and remote manipulator tool usage and control.
Thomas Lindsay, "Teleprogramming: Remote Site Research Issues: (Dissertation Proposal)", . January 1992.
Date Posted: 08 August 2007