Identification of Continuous Time Dynamical Systems with Unknown Noise Covariance.
Abstract
The present dissertation is a study of identifying parameters of a continuous-time dynamical system with noisy observation and with or without noise in the state of the system. In identifying parameters of a continuous-time dynamical system, the difficulty arises when the observation noise covariance is unknown. The present paper solves this problem in the case of a linear time invariant system with white noise affecting additively both the state and the observation. Likelihood functional cannot be obtained when the observation noise covariance is unknown. A similar procedure, however, works and the estimates are obtained by finding roots of an appropriate functional. It is shown that the estimates obtained are weakly consistent. In the special case of no noise in the state, it is further shown that similar procedure yields estimates that are strongly consistent. Consistency is proved under certain sufficient condition called the 'Identifiability Condition'. This condition is studied in detail and computational algorithm for determining the estimates is discussed.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1974
- Accession Number
- ADA005885
Entities
People
- Arunabha Bagchi
Organizations
- University of California, Los Angeles