Proceedings Article | 14 February 1992
KEYWORDS: Feedback loops, Mobile robots, Sensors, Control systems, Robotics, Complex systems, Data modeling, Human-machine interfaces, Data communications, Servomechanisms
There are two themes addressed in this paper: (1) the contrast between hierarchical and feedback-based approaches to designing robot systems, and (2) the need for an efficient human-machine interface. In the rapid pursuit of automation it is sometimes overlooked that an elaborate human-machine interface is quite necessary, despite the fact that a fully automated system, by definition, would not require much human interaction. In the future, real-time sensing, intelligent processors, and dextrous manipulators will become more viable, but until then it is necessary for human operators to provide the intelligent control functions. It is not obvious, however, how automated subsystems could account for human intervention, especially if a philosophy of pure automation dominates the design process. Teleoperation, on the other hand, emphasizes the creation of fast hardware pathways (e.g., hand controllers, exoskeletons) to communicate low-level control data to various mechanisms, while providing sensory feedback in a format suitable for human consumption (e.g., stereo display, force reflection), leaving the `intelligent' control functions to the human. These differences make it difficult to tie automation together with teleoperation while providing for graceful transitions at the appropriate times. The analysis of these issues has come of late under the title: supervisory control. Supervisory control systems must deal with various types of intelligence distributed throughout the layers of control. Typical layers are real-time servo control, off-line planning and reasoning subsystems and finally, the human operator. Design methodologies must account for the fact that the majority of the intelligence will reside with the human operator. This paper focuses on hierarchical decompositions and feedback loops as conceptual building blocks that provide a common ground for man-machine interaction. Several examples and candidate computer architectures are also discussed.
