Hybrid Control of a Robotic Manufacturing System
X.D. Koutsoukos and P.J. Antsaklis
In Proceedings of the 7th Mediterranean Conference on
Control and Automation,
pp. 144-159, Haifa, Israel. June 1999.
Abstract -- In this paper, a new approach for control of hybrid
systems is introduced and illustrated using a robotic manufacturing system.
Hybrid systems, which are used to model the physical process, and their
controllers are viewed as system components of an intelligent control
framework and they are modeled as set-dynamical systems. The central
concept studied in hybrid system modeling is quasideterminism and it is
used to address the problems that arise because of the nondeterministic
nature of the discrete approximations of the continuous dynamics.
Decision algorithms are derived based on supervisory control of
discrete-event systems described by Petri nets. The notions of
abstraction and multirate time scales play an important role in the design
of decision algorithms to supervise the operation of the system.
[postscript file] [pdf file]