Image-based autonomous landing of unmanned aerial vehicles
Unmanned aerial vehicles (UAVs) are useful in various applications, one of which is in the field of logistics, used by e-commerce companies such as Alibaba, JD.com and Amazon. However, a limited battery capacity and an insufficiently accurate navigation satellite signal are the primary factors imped...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/78744 |
| _version_ | 1811683042227388416 |
|---|---|
| author | Yong, Jian Wen |
| author2 | Low Kin Huat |
| author_facet | Low Kin Huat Yong, Jian Wen |
| author_sort | Yong, Jian Wen |
| collection | NTU |
| description | Unmanned aerial vehicles (UAVs) are useful in various applications, one of which is in the field of logistics, used by e-commerce companies such as Alibaba, JD.com and Amazon. However, a limited battery capacity and an insufficiently accurate navigation satellite signal are the primary factors impeding a more widespread adoption of this technology. Therefore, this project aims to develop a vision-based autonomous landing system of UAVs that can help to overcome these problems. There are two available approaches: position-based visual servoing (PBVS) and image based-visual servoing (IBVS). In this project, IBVS was selected for use due to its lower consumption of computational power and no need for precise camera calibration compared to PBVS. In addition, in contrast to PBVS, IBVS does not require an accurate estimation of the 3D parameters and it is more robust to errors in estimation. This report will discuss the reasons of choosing ArUco markers over the other types of fiducial markers. Besides that, the algorithms used for detecting the ArUco markers, written in MATLAB, will be discussed. The steps involved in the process are image binarization, boundaries extraction, quadrilaterals detection and eventually marker identification. This report will also discuss the use of Simulink in creating an IBVS model, which will be modified at a later stage to receive and send data using a custom ROS message. Simulations were carried out in the Ubuntu environment using tools like OpenCV, ROS and Gazebo. Several experiments were conducted to test the IBVS system designed, which include hovering and landing of UAV above a marker, and tracking of a moving marker with translational as well as rotational motion. The results obtained from the experiments were then analyzed. It is believed that, by implementing the design in this project, autonomous landing of UAVs and other applications such as hovering above moving targets and horizontal visual locking can be realized. |
| first_indexed | 2024-10-01T04:06:26Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/78744 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T04:06:26Z |
| publishDate | 2019 |
| record_format | dspace |
| spelling | ntu-10356/787442023-03-04T18:38:24Z Image-based autonomous landing of unmanned aerial vehicles Yong, Jian Wen Low Kin Huat School of Mechanical and Aerospace Engineering Engineering::Mechanical engineering Unmanned aerial vehicles (UAVs) are useful in various applications, one of which is in the field of logistics, used by e-commerce companies such as Alibaba, JD.com and Amazon. However, a limited battery capacity and an insufficiently accurate navigation satellite signal are the primary factors impeding a more widespread adoption of this technology. Therefore, this project aims to develop a vision-based autonomous landing system of UAVs that can help to overcome these problems. There are two available approaches: position-based visual servoing (PBVS) and image based-visual servoing (IBVS). In this project, IBVS was selected for use due to its lower consumption of computational power and no need for precise camera calibration compared to PBVS. In addition, in contrast to PBVS, IBVS does not require an accurate estimation of the 3D parameters and it is more robust to errors in estimation. This report will discuss the reasons of choosing ArUco markers over the other types of fiducial markers. Besides that, the algorithms used for detecting the ArUco markers, written in MATLAB, will be discussed. The steps involved in the process are image binarization, boundaries extraction, quadrilaterals detection and eventually marker identification. This report will also discuss the use of Simulink in creating an IBVS model, which will be modified at a later stage to receive and send data using a custom ROS message. Simulations were carried out in the Ubuntu environment using tools like OpenCV, ROS and Gazebo. Several experiments were conducted to test the IBVS system designed, which include hovering and landing of UAV above a marker, and tracking of a moving marker with translational as well as rotational motion. The results obtained from the experiments were then analyzed. It is believed that, by implementing the design in this project, autonomous landing of UAVs and other applications such as hovering above moving targets and horizontal visual locking can be realized. Bachelor of Engineering (Mechanical Engineering) 2019-06-26T07:00:14Z 2019-06-26T07:00:14Z 2019 Final Year Project (FYP) http://hdl.handle.net/10356/78744 en Nanyang Technological University 67 p. application/pdf |
| spellingShingle | Engineering::Mechanical engineering Yong, Jian Wen Image-based autonomous landing of unmanned aerial vehicles |
| title | Image-based autonomous landing of unmanned aerial vehicles |
| title_full | Image-based autonomous landing of unmanned aerial vehicles |
| title_fullStr | Image-based autonomous landing of unmanned aerial vehicles |
| title_full_unstemmed | Image-based autonomous landing of unmanned aerial vehicles |
| title_short | Image-based autonomous landing of unmanned aerial vehicles |
| title_sort | image based autonomous landing of unmanned aerial vehicles |
| topic | Engineering::Mechanical engineering |
| url | http://hdl.handle.net/10356/78744 |
| work_keys_str_mv | AT yongjianwen imagebasedautonomouslandingofunmannedaerialvehicles |