| Summary: | Approved for public release; distribution is unlimited. === This thesis develops, implements and tests a Tactical Decision Aid for a Reactive Target ASW Active Search. The mode! uses a Bayesian Filtering Process to fuse information from a real world search conducted by several assets with information from a Monte Carlo Simulation that encompasses five hundred equally likely different possible initial positions and behaviors of the real target. A Reactive Target Model resembles the behavior of a target that is always aware and reacts because of the presence and activity of the searchers. An initial 'prior', or best estimate of the location of the target is updated using the movement of the simulated targets, the negative information conveyed in an unsuccessful search over a period of time and the positive information implied in a contact report. The search effort is measured using a Fixed Scan Stochastic Model that solves the Sonar Equation limited by noise and reverberation. As a result of updating the prior, a 'posterior' distribution is obtained. The Law of Total Probabilities is used to render a probability map of the location of the Target by mapping color intensities to probabilities. A recursive expression for evaluating a contact report is also developed.
|
|---|
