Adaptive Control of Linear Stochastic Systems.
TEXAS UNIV AUSTIN ELECTRONICS RESEARCH CENTER
Pagination or Media Count:
The problem of controlling unknown stochastic systems stems from the need that in many practical design situations, the engineer does not have at his disposal sufficient data to derive a complete mathematical model of the process he endeavors to control. Instead, uncertainty may exist with respect to certain parameters in the system description. In this study, the unknown parameters in the system are assumed to be constant and are modeled as random variables. Bayesian approach is used, in that it is assumed that the a-priori probability distribution for the unknown parameters is available. The adaptive-control algorithm developed in this research was used to study economic stabilization for a short-term economic model of the post-Korean War U. S. economy. This is a linear 28-state variable model with three control variables. The experimental results demonstrate that the adaptive-control approach is valuable as a tool for policy planning and is helpful in understanding the dynamic structure of econometric models. Modified author abstract
- Economics and Cost Analysis
- Statistics and Probability