Dynamic diagnostic and decision procedures under uncertainty
In this paper, we consider uncertainty that arises when the true state x {element_of} E is not accessible to direct observation and remains unknown. Instead, we observe some features {theta} {element_of} {Theta} that carry a certain information about the true state. This information is described by the conditional distribution P({Theta}{vert_bar}E), which we call the linkage distribution. Regarding this distribution we assume that it exists but is unknown. This leads to uncertainty with respect to states from E and the linkage distribution P({Theta}{vert_bar}E), which we denote by NEP. The substantive problem can be stated as follows: from observations of the features {theta}{element_of}{Theta} made at each time instant n = 1,2,...,recognize the state x {element_of} E, identify the linkage distribution P, and use the results of recognition and identification to choose a decision y {element_of} Y so that the decision process is optimal in some sense. State recognition is the subject of diagnostics. The uncertainty NEP thus generates a problem of diagnostics and dynamic decision making.
- OSTI ID:
- 441168
- Journal Information:
- Cybernetics and Systems Analysis, Vol. 30, Issue 3; Other Information: PBD: Jan 1995; TN: Translated from Kibernetika i Sistemnyi Analiz; No. 3, 87-104(May-Jun 1994)
- Country of Publication:
- United States
- Language:
- English
Similar Records
Multiple sensor fusion under unknown distributions
A nonlinear regression estimate