Variable Metric Methods and Filtering Theory
In this paper we show that there is a close relationship between variable metric methods of function minimization and filtering of linear stochastic systems with disturbances which are modelled as unknown but bounded functions. We develop new variable metric algorithms for function minimization.
KeywordsSupport Function Function Minimization Rapid Convergence Filter Theory Derivative Computation
Unable to display preview. Download preview PDF.
- D. Bertsekas, “Control of Uncertain Systems with a Set Membership Description of Uncertainty,” Ph.D. Thesis, Department of Electrical Engineering, M.I.T., 1971.Google Scholar
- S. K. Mitter and P. Toldalagi, to appear.Google Scholar
- M.J.D. Powell, “Convergence Properties of a Class of Minimization Algorithms, in Nonlinear Programming, Vol. 2, editors Mangasarian, Meyer and Robinson, McGraw-Hill, New York, 1975.Google Scholar
- S. W. Thomas, “Sequential Estimation Techniques for Quasi-Newton Algorithms, Cornell University, Department of Computer Science, Technical Report TR75-227, 1975.Google Scholar