A.R Drone vision-guided searching

Long, Derek (2013) A.R Drone vision-guided searching. [USQ Project]

[img]
Preview
Text
Long_2013.pdf

Download (19Mb) | Preview

Abstract

UAVs are becoming more and more capable of performing complex tasks in today's world. Military and scientific applications are becoming increasingly less dependant on
human monitoring, with the increased capabilities of UAV sensors and control abilities. There is also an increasing need for automation of repetitive tasks in industries around
the world. There is amazing opportunity for process efficiency improvement if a machine can be substituted in place of a person to perform a menial tasks.

This dissertation is aimed at getting a UAV to perform an automated function for use in monitoring or search and rescue. It proposes a method for an A.R. Drone to home
in on a object 'beacon' in a set search area. Having UAVs automated in this way has the potential to both improve monitoring process that already exist, and open up new
opportunities for aerial applications that were previously not possible.

The original implementation of this homing application achieved the base function only. The drone was able to search a small area for a coloured ball, and navigate to it,
once the ball had been located. Several tests were performed in preparation for this, developing machine vision and tracking functions that would be needed for the drone to perform such a task.

Work at this stage is being done to further investigate how to enhance the abilities and efficiency of the algorithms that form the foundation of the application. Detection
robustness and control smoothing are priorities in improvement, as well as developing additional features, such as obstacle avoidance.


Statistics for USQ ePrint 24666
Statistics for this ePrint Item
Item Type: USQ Project
Item Status: Live Archive
Faculty / Department / School: Current - Faculty of Health, Engineering and Sciences - School of Mechanical and Electrical Engineering
Supervisors: Low, Tobias; McCarthy, Cheryl
Date Deposited: 27 Feb 2014 04:46
Last Modified: 27 Feb 2014 04:46
Uncontrolled Keywords: ar drone; vision searching; uav
Fields of Research : 09 Engineering > 0901 Aerospace Engineering > 090106 Flight Dynamics
URI: http://eprints.usq.edu.au/id/eprint/24666

Actions (login required)

View Item Archive Repository Staff Only