Award Date

5-1-2014

Degree Type

Thesis

Degree Name

Master of Science in Computer Science

Department

Computer Science

First Committee Member

Wolfgang Bein

Second Committee Member

Kazem Taghva

Third Committee Member

Laxmi P. Gewali

Fourth Committee Member

Emma Regentova

Number of Pages

74

Abstract

The most common approach for border patrol operations is the use of human personnel and manned ground vehicles, which is expensive, at times inefficient and sometimes even hazardous to people involved. The length of the US border, mostly covering unpopulated areas, with harsh atmospheric conditions makes it more susceptible to illegal human activities. Automated border surveillance by unattended, fixed, ground sensors forming an electronic fence has proven expensive, inefficient and was prone to unacceptable rate of false alarms.

A better approach would be using Unmanned Aerial Vehicles (UAVs) in combination with such ground sensors. This would help improve the overall effectiveness of the surveillance system as a UAV could first scan the alert area before sending in personnel and vehicles, if deemed necessary.

In this thesis, we are proposing border surveillance using multiple Unmanned Aerial Vehicles (UAVs) in combination with alert stations consisting of Unattended Ground Sensors (UGSs) along the border line or fence. Upon detecting an event, an alert would be triggered by any UGS. We simulate this process by reading probability data for different timestamps from a text file. And, based on utility values of each stations, two UAVs decide on which alert stations to service.

Keywords

Aerial surveillance; Border patrols; Computer algorithms; Drone aircraft; U.S. Border Patrol

Disciplines

Computer Sciences

Language

English


Share

COinS