Abstract
Machine identification of discrete event systems (DES) addresses the issue of identifying an unknown system based on an externally observed sample path from the unknown system. Online Modeling Refinement studies the continuing machine identification process when the observed sample path is updated incrementally. The notion of information embedded in a sample path is defined. By taking advantage of the structural similarity between successive observed sample paths, the computational requirement for the proposed online modeling refinement algorithm is kept minimal. An example is provided to show how identification results converge as the incrementally observed sequence is accumulated over time.
Key words:
Notes
Corresponding author. (Tel: 886–2–27376710; Email: [email protected])