The Complementary Model in Continuous/Discrete Smoothing
We consider the problem of smoothing a continuous-time random process using irregularly spaced noisy samples. If the random process to be smoothed is generated by a linear state model, the relevant Hamiltonian system can be easily derived using the concept of complementary model, introduced by Weinert and Desai . All smoothing algorithms can then be obtained via various changes of variables in the Hamiltonian system.
KeywordsHamiltonian System Smoothing Algorithm Linear State Model Complementary Model Noisy Sample
Unable to display preview. Download preview PDF.
- H.L. Weinert, “On adjoint and complementary systems,” Proc. IEEE Conf. Decision and Control, 1981, 123–124.Google Scholar
- M.G. Bello, “Centralized and decentralized map updating and terrain masking analysis,” Ph.D. Dissertation, Elect. Engr. Comp. Sci. Dept., M.I.T., 1981 (Report LIDS-TH-1123).Google Scholar
- M.B. Adams, A.S. Willsky and B.C. Levy, “Linear estimation of boundary value stochastic processes,” IEEE Trans. Auto. Contr., to appear.Google Scholar
- A.E. Bryson and Y.C. Ho, Applied Optimal Control, Wiley, New York, 1975.Google Scholar
- H.L. Weinert, “Sample function properties of a class of smoothed estimates,” IEEE Trans. Auto. Contr., vol. AC-28, 1983.Google Scholar
- R. Kohn and C.F. Ansley, “On the smoothness properties of the best linear unbiased estimate of a stochastic process observed with noise,” submitted to Ann. Stat., 1982.Google Scholar