Stochastic Line Search Using UUVs

Abstract

Unmanned underwater vehicles (UUVs) are increasingly being used in a diverse range of applications. In one particular application, we analyze UUV operations for location, detection and classification of mines. The mission objective is to search the area of interest, using underwater imaging sensors such as side scan sonars, until either the first mine is located, or it is verified that none can be found. Communication constraints require that the vehicle be connected with physically for downloads. In such as scenario, the search area can be considered as a line, and prior probabilities of finding a mine on the line can be related to external considerations such as the bottom characteristics, etc. The optimization problem is the determination of a sequence of points on the line where the UUV should be configured to return for a data download, so as to minimize the expected mission time. Operational models are defined and analytical expressions and numerical results describe the optimal strategies for searching with several distributions and return point specifications