Object-Oriented Modeling: Tools and Techniques for Capturing Properties of Physical Systems

François E. Cellier
Institute of Computational Science
ETH Zürich
CH-8092 Zürich
Switzerland
Email: FCellier@Inf.ETHZ.CH
Bernard P. Zeigler
Arizona Center for Integrative Modeling & Simulation
Department of Electrical and Computer Engineering
The University of Arizona
Tucson, Arizona 85721
U.S.A.
Email: Zeigler@ECE.Arizona.Edu
Andrew H. Cutler
Space Engineering Research Center
The University of Arizona
Tucson, Arizona 85721
U.S.A.

Keywords

Artificial Intelligence
Decentralized Control
Event-Based Control
Failure Detection
Intelligent Machines
Knowledge Abstraction
Model Synthesis
Object-Oriented Modeling
Process Control
Time-Windows

Abstract

Mathematical modeling means the formal encoding of knowledge about a dynamical system. Knowledge can be grouped into behavioral knowledge and structural knowledge. Behavioral knowledge is local knowledge relating to a particular experiment applied to a system or model. Behavioral knowledge is what is generated in a real-world experiment or during a simulation run. Structural knowledge is global knowledge relating to a system or model, irrespective of the experiment that is performed. A model is a formal encoding of structural knowledge of a system. Structural knowledge can be further decomposed into functional knowledge, coupling knowledge, decomposition knowledge, and taxonomic knowledge. In this paper, a methodology is presented that helps to organize the structural knowledge about a system to be described. It enables to encode separately and in an organized fashion functional, coupling, decomposition, and taxonomic knowledge about a system. The methodology lends itself to the implementation of automated procedures for deductive as well as inductive model synthesis necessary for the realization of high-autonomy intelligent control systems. A fairly involved example concludes the paper.

Interested in reading the full paper? (10 pages, 937,755 bytes, pdf)

Homepage