Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Coverage path planning for autonomous surveying of plant species using unmanned aerial vehicles (UAVs)
Thesis (MEng)--Stellenbosch University, 2024.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
Stellenbosch : Stellenbosch University
2025
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613974188523520 |
|---|---|
| access_status_str | Open Access |
| author | Cilliers, Abraham Albertus |
| author2 | Engelbrecht, J. A. A. |
| author_browse | Cilliers, Abraham Albertus Engelbrecht, J. A. A. |
| author_facet | Engelbrecht, J. A. A. Cilliers, Abraham Albertus |
| author_sort | Cilliers, Abraham Albertus |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description |
Thesis (MEng)--Stellenbosch University, 2024. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/131621 |
| institution | Stellenbosch University (South Africa) |
| language | English |
| last_indexed | 2026-06-10T12:44:39.798Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2025 |
| publishDateRange | 2025 |
| publishDateSort | 2025 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/131621 Coverage path planning for autonomous surveying of plant species using unmanned aerial vehicles (UAVs) Cilliers, Abraham Albertus Engelbrecht, J. A. A. Stellenbosch University. Faculty of Engineering. Dept. of Electrical and Electronic Engineering. Autonomous vehicles Unmanned aerial vehicles Robots -- Navigation UCTD Thesis (MEng)--Stellenbosch University, 2024. ENGLISH ABSTRACT: This thesis presents the development of a coverage path planning system for autonomously surveying designated areas, using one or more unmanned aerial vehicles with camera payloads so that the captured images can be used to identify plants belonging to target species. Image processing techniques are used to detect and segment areas of vegetation in satellite images to identify the areas that must be covered. The segmented map is then discretised into a search grid, where the grid cell size is determined by the ground sampling distance needed to ensure a resolution that is high enough to detect individual plants. A discrete search grid is overlayed on the satellite image and each grid cell is labelled as vegetation, empty, or obstacle. A new grid-based coverage path planning algorithm was developed to efficiently cover the potentially disconnected grid cells containing vegetation while avoiding the grid cells containing obstacles. The new coverage path planning algorithm is based on the well-known A* path-finding algorithm, but extends it to perform coverage path planning by introducing a novel reward function that maximises the vegetation exploration rate. The coverage path planning algorithm takes the limited fuel or battery capacity of the UAV into account, by placing a constraint on the total number of grid cells that can be visited. The coverage path plans are then converted to GPS waypoints which can be uploaded to the UAVs using the mission planning software on the UAV ground station. The UAVs then autonomously execute the required flight paths and take photos at the specified locations. The system was tested in simulation and with practical flight tests using a physical UAV at a real-world location. Both the simulation results and the practical flight test results showed that the UAV can accurately execute the planned coverage paths. The flight tests proved that the system can survey disconnected vegetation areas, navigate complex environments containing obstacles and no-fly zones, and perform consecutive path generation to enable refuelling or the use of multiple UAVs. The results also show that our proposed system covers the required areas of vegetation more efficiently (travelling shorter distances) than the standard lawnmower patterns provided by commercial systems. AFRIKAANSE OPSOMMING: Hierdie tesis beskryf die ontwikkeling van ’n dekkingspadbeplanningstelsel wat ’n gegewe area outonoom sal dek met behulp van een of meer onbemande vliegtuie met kameraloonvragte, sodat die beelde gebruik kan word om plante wat aan teikenspesies behoort, te identifiseer. Beeldverwerkingstegnieke word gebruik om areas van plantegroei in satellietbeelde te identifiseer en te segmenteer, om die areas wat gedek moet word, af te baken. Die gesegmenteerde kaart word dan gediskretiseer in ’n soekrooster, waar die roosterselgrootte bepaal word deur die grondmonsternemingsafstand wat nodig is om ’n resolusie te verseker wat hoog genoeg is om individuele plante te kan herken. Die diskrete soekrooster word op die satellietbeeld oorgelˆe en elke roostersel is gemerk as plantegroei, leeg of obstruksie. ’n Nuwe roostergebaseerde dekkingspadbeplanningsalgoritme is ontwikkel om die potensieel verwyderde roosterselle wat plantegroei bevat doeltreffend te dek, terwyl die roosterselle wat obstruksies bevat, vermy word. Die nuwe dekkingspadbeplanningsalgoritme is gebaseer op die bekende A* padvindalgoritme, maar brei dit uit om dekkingspadbeplanning uit te voer deur ’n nuwe beloningsfunksie in te stel wat die plantegroei ontdekkingtempo maksimeer. Die dekkingspadbeplanningsalgoritme neem die beperkte brandstof of batterykapasiteit van die UAV in ag deur ’n beperking te plaas op die totale aantal roosterselle wat besoek kan word. Die dekkingspadplanne word dan omgeskakel na GPS wegpunte wat na die UAV’s opgelaai kan word deur die missiebeplanningsagteware op die UAV grondstasie te gebruik. Die UAV’s voer dan outonoom die vereiste vlugpaaie uit en neem foto’s op die gespesifiseerde plekke. Die stelsel is getoets beide in simulasie en met praktiese vlugtoetse met ’n fisiese onbemande vliegtuig in ’n werklike ligging. Beide die simulasie en die praktiese vlugtoetsresultate het getoon dat die UAV die beplande dekkingspaaie, akkuraat kon uitvoer. Die vlugtoetse het bewys dat die stelsel in staat was om verwyderde soekgebiede te dek, komplekse omgewings met obstruksies of geen-vlieg sones te navigeer, en opeenvolgende padgenerering uit te voer om hervulling of die gebruik van veelvuldige UAV’s moontlik te maak. Die resultate wys ook dat ons voorgestelde stelsel die nodige areas van plantegroei meer doeltreffend dek (deur korter afstande af te lˆe) as die meer standaard grassnyer patrone wat verskaf word deur kommersiele stelsels. Masters 2025-01-29T12:37:50Z 2025-01-29T12:37:50Z 2024-12 Thesis https://scholar.sun.ac.za/handle/10019.1/131621 en Stellenbosch University xiv, 105 pages : illustrations application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Autonomous vehicles Unmanned aerial vehicles Robots -- Navigation UCTD Cilliers, Abraham Albertus Coverage path planning for autonomous surveying of plant species using unmanned aerial vehicles (UAVs) |
| title | Coverage path planning for autonomous surveying of plant species using unmanned aerial vehicles (UAVs) |
| title_full | Coverage path planning for autonomous surveying of plant species using unmanned aerial vehicles (UAVs) |
| title_fullStr | Coverage path planning for autonomous surveying of plant species using unmanned aerial vehicles (UAVs) |
| title_full_unstemmed | Coverage path planning for autonomous surveying of plant species using unmanned aerial vehicles (UAVs) |
| title_short | Coverage path planning for autonomous surveying of plant species using unmanned aerial vehicles (UAVs) |
| title_sort | coverage path planning for autonomous surveying of plant species using unmanned aerial vehicles uavs |
| topic | Autonomous vehicles Unmanned aerial vehicles Robots -- Navigation UCTD |
| url | https://scholar.sun.ac.za/handle/10019.1/131621 |
| work_keys_str_mv | AT cilliersabrahamalbertus coveragepathplanningforautonomoussurveyingofplantspeciesusingunmannedaerialvehiclesuavs |