Techniques for Modeling Stochastics Dynamical Systems.
Final rept. 4 Feb 76-1 Jan 83,
HARVARD UNIV CAMBRIDGE MA DIV OF APPLIED SCIENCES
Pagination or Media Count:
This report summarizes the principle research findings on 30 published papers and 3 Ph.D. theses completed with partial support of the Army Research Office. Nonlinear stochastic systems arise in various areas of engineering including communications, vibrations and control. Under the grant and contract referenced above the author undertook to produce rich class models which were, at the same time, analytically tractable. He studied these questions with respect to computing the statistical properties of solutions and with respect to finding sufficient statistics for nonlinear estimation problems. Also investigated were the specific questions which arise in an attempt to adaptively control systems via system identification and subsequent control.
- Statistics and Probability