- Title
- A vision based multirotor aircraft for use in the security industry
- Creator
- Nelson, Benjamin David
- Subject
- Drone aircraft -- South Africa
- Subject
- Mechatronics -- Research
- Date Issued
- 2020
- Date
- 2020
- Type
- Thesis
- Type
- Masters
- Type
- MEng
- Identifier
- http://hdl.handle.net/10948/46816
- Identifier
- vital:39664
- Description
- This research consisted of developing a vision based multirotor aircraft that could be used in the security industry. A second-hand aircraft was purchased and modified. The aircraft made use of a Pixhawk flight controller and a Odroid XU4 companion computer, which resulted in the computer injecting commands into the flight controller. Robot Operating System was installed and used on the companion computer to integrate the vision system and the aircraft. The vision system was designed to help develop a landing system where the aircraft would land on an ArUco marker. The vision system also allowed the aircraft to detect and follow humans. A Software in the Loop (SITL) was run alongside Gazebo, allowing the developed landing system and the human detecting system to be simulated and tested. The developed landing system was implemented on the aircraft, where the developed landing system was tested and compared to the aircraft’s current GPS based landing system. The developed landing system obtained a better overall accuracy , while also taking longer to land the aircraft compared to the GPS based landing system. There were also numerous manual and autonomous test flights implemented on the aircraft.
- Format
- xviii, 257 leaves
- Format
- Publisher
- Nelson Mandela University
- Publisher
- Faculty of Engineering, the Built Environment and Technology
- Language
- English
- Rights
- Nelson Mandela University
- Hits: 1062
- Visitors: 1546
- Downloads: 733
Collections
NMU School of Engineering
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | SOURCE1 | Nelson, BD 214006891 Dissertation April 2020.pdf | 12 MB | Adobe Acrobat PDF | View Details Download |