Estimates conformity principle in the problems of identification

  • Authors:

  • Vladimir Fursov

  • Affiliations:

  • Image Processing Systems Institute, The Russian Academy of Sciences, Samara, Russia

  • Venue:
  • ICCS'03 Proceedings of the 2003 international conference on Computational science: PartII
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

A non-traditional approach to solving the identification problems using a small number of observations is developed. The approach is formulated as an estimate conformity principle. The problem of identifying a linear dynamic model is treated within this approach. A significant feature of the above method is the use of a quality criterion not requiring the knowledge of a priori probability noise distributions. The method is well suitable for real-time object identification in adaptive control systems.