Date of Award

9-2003

Document Type

Thesis

Degree Name

Master of Science

Department

Department of Aeronautics and Astronautics

First Advisor

David R. Jacques, PhD

Abstract

Several research simulations have been created to support development and refinement of teamed autonomous agents using decentralized cooperative control algorithms. Simulation is the necessary tool to evaluate the performance of decentralized cooperative control algorithms, however these simulations lack a method to validate their output. This research presents a method to validate the performance of a decentralized cooperative control simulation environment for an autonomous Wide Area Search Munition (WASM). Rigorous analytical methods for six wide area search and engagement scenarios involving Uniform, Normal, and Poisson distributions of N real targets and M false target objects are formulated to generate expected numbers of target attacks and kills for a searching WASM. The mean value based on the number of target attack and kills from Monte Carlo simulations representative of the individual scenarios are compared to the analytically derived expected values. Emphasis is placed on Wide Area Search Munitions (WASMs) operating in a multiple target environment where a percentage of the total targets are either false targets or may be misconstrued as false by varying the capability of the WASM’s Automatic Target Recognition (ATR) capability.

AFIT Designator

AFIT-GAE-ENY-03S-14

DTIC Accession Number

ADA420800

Download

Share

COinS