The scientific inferential relationships between statistical estimation, decision theory, and modern filter theory.
Scientific inferential relationships between statistical estimation, decision theory and modern filter theory
SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Scientific inferential relationships between statistical estimation, decision theory and modern filter theory
Heuristics that guide search are critical when solving large planning and scheduling problems, but most variable and value ordering heuristics are sensitive to only one feature of the search state. One wants to combine evidence from all features of the search state into a subjective probability that a value choice is best, but there has been no solid semantics for merging evidence when it is conceived in these terms. Instead, variable and value ordering decisions should be viewed as problems in decision theory. This led to two key insights: (1) The fundamental concept that allows heuristic evidence to be merged is the net incremental utility that will be achieved by assigning a value to a variable. Probability distributions about net incremental utility can merge evidence from the utility function, binary constraints, resource constraints, and other problem features. The subjective probability that a value is the best choice is then derived from probability distributions about net incremental utility. (2) The methods used for rumor control in Bayesian Networks are the primary way to prevent cycling in the computation of probable net incremental utility. These insights lead to semantically justifiable ways to compute heuristic variable and value ordering decisions that merge evidence from all available features of the search state.
Learning technique incorporating feature selection and decision theory for finite memory nonlinear process identification
Human behavior is a complex nature which depends on circumstances and decisions varying from time to time as well as place to place. The way a decision is made either directly or indirectly related to the availability of the options. These options though appear at random nature, have a solid directional way for decision making. In this paper, a decision theory is proposed which is based on human behavior. The theory is structured with model sets that will show the all possible combinations for making a decision, A virtual and simulated environment is considered to show the results of the proposed decision theory
Decision theory and extraterrestrial resource utilization in space mission planning
The primary theme of this investigation is a decision theoretic account of conditional ought statements (e.g., 'You ought to do A, if C') that rectifies glaring deficiencies in classical deontic logic. The resulting account forms a sound basis for qualitative decision theory, thus providing a framework for qualitative planning under uncertainty. In particular, we show that adding causal relationships (in the form of a single graph) as part of an epistemic state is sufficient to facilitate the analysis of action sequences, their consequences, their interaction with observations, their expected utilities, and the synthesis of plans and strategies under uncertainty.
Statistical decision theory applied to Martian atmosphere analysis, life detection experiments, and gas chromatogram measurements of n-alkane distributions in materials
Some fundamentals of reliability analysis as applicable to aerospace structures are reviewed, and the concept of a test option is introduced. A decision methodology, based on statistical decision theory, is developed for determining the most cost-effective design factor and method of testing for a given structural assembly. The method is applied to several Saturn V and Space Shuttle structural assemblies as examples. It is observed that the cost and weight features of the design have a significant effect on the optimum decision.
Measured values of several variables are incorporated into the calculation of snow water equivalent as measured from an aircraft by snow attenuation of terrestrial gamma radiation. Bayesian decision theory provides a snow water equivalent measurement by taking into account the uncertainties in the individual measurement variables and filtering information about the measurement variables through prior notions of what the calculated variable (water equivalent) should be.
This chapter addresses the problem of using redundancy that exists between dissimilar systems in aircraft flight control. That is, using the redundancy that exists between a rate gyro and an accelerometer—devices that have dissimilar outputs which are related only through the dynamics of the aircraft motion. Management of this type of redundancy requires advanced logic so that the system can monitor failure status and can reconfigure itself in the event of one or more failures. In this chapter an optimal decision theory is tutorially developed for the management of sensor redundancy and the theory is applied to two aircraft examples. The first example is the space shuttle and the second is a highly maneuvering high performance aircraft—the F8-C. The examples illustrate the redundancy management design process and the performance of the algorithms presented in failure detection and control law reconfiguration.
Explore the source record for details and available documents.
In this document, we describe a simple autonomous star identification algorithm which is effective using a narrow field of view (2 degrees), making the use of a science camera for star identification feasible.
The focus of this report is the consideration of one form of cognition, judgment and decision making, while examining some information management issues associated with the implementation of new forms of automation. As technology matures and more tasks become suitable to automation, human factors researchers will have to consider the effect that increasing automation will have on operator performance. Current technology allows flight deck designers the opportunity to automate activities involving substantially more cognitive processing.
Decision theory and binary detection of signal transmission in noisy channel, and determination of noise caused by tape in magnetic recording system
Decision making, communication theory, and space communications
Personnel decision theory deduced from human as tracking operator
Quantum statistical decision theory for detection and estimation of signals in random noise
Theory of group decision behavior tested on dyads