10.2514/6.2018-1894">
 

A Utility Approach to UAS-Based Persistent ISR

Document Type

Conference Proceeding

Publication Date

1-8-2018

Abstract

The Persistent Intelligence, Surveillance, and Reconnaissance problem seeks to provide timely collection and delivery of data from a set of prioritized geographic points of interest (tasks) using an autonomous Unmanned Aerial System. In the literature, this problem is often called persistent monitoring, which is a type of Vehicle Routing Problem. The objective in PISR is to repeatedly visit tasks, preferably in an efficient manner, where efficiency is usually measured by the revisit period of tasks. Past approaches to PISR have either been based on combinatorial optimization or utility theory. Combinatorial optimization solutions can produce locally optimal results, but have drawbacks that make them difficult to implement. These include high processing overhead, the use of complex linear program- ming algorithms, the need for stable communication links, and computational complexity that increases with the number of tasks and vehicles. For practical PISR, a utility function approach avoids many of the combinatorial difficulties. This paper studies the merits and challenges of a new utility function, the Maximal Distance Discounted and Weighted Revisit Period (MD2WRP), and sets out to characterize its behavior using a set of toy problems. A plan is then described to measure performance of the proposed utility function, tune its parameters, and compare it to other PISR strategies.

Comments

This conference paper is available from the publisher, AIAA, through subscription or purchase using the DOI link below.

Author note: Christopher Olsen was an AFIT PhD candidate at the time of this conference. (AFIT-ENY-DS-18-S-067, September 2018)

Source Publication

2018 AIAA Information Systems-AIAA Infotech @ Aerospace

This document is currently not available here.

Share

COinS