Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Cooperative search and rescue using a scheduling algorithm
Thesis (MEng)--Stellenbosch University, 2025.
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_ | 1867614112303808512 |
|---|---|
| access_status_str | Open Access |
| author | Buys, Stephano |
| author2 | Schoeman, J. C. |
| author_browse | Buys, Stephano Schoeman, J. C. |
| author_facet | Schoeman, J. C. Buys, Stephano |
| author_sort | Buys, Stephano |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Thesis (MEng)--Stellenbosch University, 2025. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/134556 |
| institution | Stellenbosch University (South Africa) |
| language | English |
| last_indexed | 2026-06-10T12:46:51.765Z |
| 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:10019.1/134556 Cooperative search and rescue using a scheduling algorithm Buys, Stephano Schoeman, J. C. Engelbrecht, J. A. A. (Japie) Stellenbosch University. Faculty of Engineering. Dept. of Electrical & Electronic Engineering. Search and rescue operations Time-sharing computer systems Scheduling -- Data processing Multiagent systems Thesis (MEng)--Stellenbosch University, 2025. Buys, S. 2025. Cooperative search and rescue using a scheduling algorithm. Unpublished masters thesis. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/e8c0f195-7216-4f5e-9d5b-c49a663b7c39 ENGLISH ABSTRACT: Autonomous search and rescue (SAR) operations demand innovative solutions that can efficiently coordinate multiple search agents in complex and dynamic environments. The autonomous SAR problem is often modelled as a coverage path planning (CPP) problem, as both aim to systematically explore all points in a given area while operating under the critical time pressures inherent in SAR scenarios. Many existing methods divide the search environment into distinct sub-areas to enhance search efficiency for multiple search agents. However, the strict constraints imposed to achieve optimal performance in these settings often limit the algorithm’s adaptability for real-world scenarios. Two typical real-world scenarios include limited fuel availability and partially mapped or fully unmapped environments. Limited fuel availability often necessitates reliance on a single refuelling station, making such divisions impractical. Additionally, partially mapped environments complicate the process of optimally dividing the search area. This thesis presents a novel framework for search and rescue missions that integrates a scheduler and a multi-agent path planner. The scheduler allocates exploration cells, or frontiers, to agents using a greedy-heuristic strategy, while the multi-agent path planner ensures collision-free trajectories, enabling effective collaboration between agents. The approach was further extended to operate in unmapped environments, where agents adapt to unknown terrain during execution, and in scenarios constrained by limited fuel, incorporating refuelling strategies to maintain mission continuity. The proposed system was tested in simulations of varying complexity, including real-world-inspired scenarios, to evaluate metrics such as success rate, computational scalability, and the efficiency of the planned trajectories. The results demonstrate the system’s ability to outperform baseline approaches, achieving higher success rates for a wide range of environmental types. These findings underline the framework’s potential as a practical tool for autonomous search and rescue missions, advancing the capabilities of UAV systems in critical applications. AFRIKAANSE OPSOMMING: Outonome soek-en-redding-operasies vereis innoverende oplossings wat verskeie onbemande lugvoertuie doeltreffend kan ko¨ordineer in komplekse en dinamiese omgewings. Die outonomesoek-en-redding-probleem word dikwels gemodelleer as ’n dekkingroetebeplanning probleem, aangesien beide daarop gemik is om alle punte in ’n gegewe area sistematies te verken terwyl hulle onder die dringende tydsdruk werk wat inherent is aan soek-en-redding-operasies. Baie bestaande metodes verdeel die soekomgewing in afsonderlike subareas om soekdoeltreffendheid vir verskeie soekagente te verbeter. Die streng beperkings wat egter opgelˆe word om optimale prestasie in hierdie instellings te bereik, beperk dikwels die algoritme se aanpasbaarheid in werklike operasies. Twee tipiese werklike omstandighede sluit beperkte brandstofbeskikbaarheid en gedeeltelik gekarteerde of heeltemal ongekarteerde omgewings in. Beperkte brandstofbeskikbaarheid vereis dikwels afhanklikheid van ’n enkele brandstofstasie, wat sulke verdelings in the omgewing onprakties maak. Verder bemoeilik gedeeltelik gekarteerde omgewings die proses om die soekarea optimaal te verdeel. Hierdie tesis stel ’n nuwe raamwerk vir soek- en reddingsmissies voor wat ’n skeduleerder en’n multi-agentroetebeplanner integreer. Die skeduleerder ken selle toe aan agente deur gebruik te maak van ’n gulsige-heuristiese strategie, terwyl die multi-agentroetebeplanner botsingsvrye trajekte verseker, wat effektiewe samewerking tussen agente moontlik maak. Die benadering is verder uitgebrei om in ongeskepte omgewings te funksioneer, waar agente tydens uitvoering by onbekende terrein aanpas, en in situasies met beperkte brandstof, wat hervullingstrategie¨e insluit om missiekontinu¨ıteit te handhaaf. Die voorgestelde stelsel is in simulasies van verskillende kompleksiteit getoets, insluitend simulasies ge¨ınspireer deur werklike situasies, om maatstawwe soos sukseskoers, berekenkundigsskaalbaarheid, en die effektiwiteit van die beplande trajekte te evalueer. Die resultate toon dat die stelsel die vermo¨e het om ander benaderings te oortref, met ho¨er sukseskoerse oor ’n wye reeks omgewingstipes. Hierdie bevindinge beklemtoon die raamwerk se potensiaal as ’n praktiese hulpmiddel vir outonome soek-en-redding-missies, wat die vermo¨ens van UAV-stelsels in kritieke toepassings verbeter. Masters 2025-12-12T11:46:24Z 2025-12-12T11:46:24Z 2025-12 Thesis https://scholar.sun.ac.za/handle/10019.1/134556 en Stellenbosch University viii, 75 pages : illustrations application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Search and rescue operations Time-sharing computer systems Scheduling -- Data processing Multiagent systems Buys, Stephano Cooperative search and rescue using a scheduling algorithm |
| title | Cooperative search and rescue using a scheduling algorithm |
| title_full | Cooperative search and rescue using a scheduling algorithm |
| title_fullStr | Cooperative search and rescue using a scheduling algorithm |
| title_full_unstemmed | Cooperative search and rescue using a scheduling algorithm |
| title_short | Cooperative search and rescue using a scheduling algorithm |
| title_sort | cooperative search and rescue using a scheduling algorithm |
| topic | Search and rescue operations Time-sharing computer systems Scheduling -- Data processing Multiagent systems |
| url | https://scholar.sun.ac.za/handle/10019.1/134556 |
| work_keys_str_mv | AT buysstephano cooperativesearchandrescueusingaschedulingalgorithm |