On-line determination of time-delay using multiple recursive estimators and fuzzy reasoning

• Published in: International Conference on Control '94 pp. 1436 - 1441
• Main Authors: Readle, J.C, Henry, R.M
• Format: Conference Proceeding
• Language: English
• Published: London IEE 1994
• Subjects: Artificial intelligence; Formal logic; Interpolation and function approximation (numerical analysis); Self-adjusting control systems; Simulation, modelling and identification; least squares approximations; identification; delays; fuzzy logic; multiple recursive estimators; adaptive control; uncertainty handling; fuzzy reasoning; recursive least squares; inference mechanisms; hybrid time delay estimation; blackboard based online identification
• ISBN: 0852966105; 9780852966105
• DOI: 10.1049/cp:19940348

A hybrid time-delay estimation technique is proposed, which uses both existing algorithmic techniques and fuzzy logic based reasoning. The two algorithmic techniques used are a single high order recursive least squares (RLS) estimator and multiple second order RLS estimators each with a different integer time-delay. The paper first describes the various techniques used for time-delay estimation and justifies the approach used in this paper. The operation and development of the technique is then briefly described. Examples of simulated test cases to assess the technique are described and discussed. It is concluded that the approach proposed provides a reliable and robust technique suitable for practical application. Some extensions to the approach are presented including determination of noninteger time-delays.