Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Map point selection for hardware constrained visual simultaneous localisation and mapping.
Thesis (PhD)--Stellenbosch University, 2024.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | en_ZA en_ZA |
| Published: |
Stellenbosch : Stellenbosch University
2024
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613898944806912 |
|---|---|
| access_status_str | Open Access |
| author | Muller, Christiaan |
| author2 | Van Daalen, Corne E.
|
| author_browse | Muller, Christiaan Van Daalen, Corne E. |
| author_facet | Van Daalen, Corne E.
Muller, Christiaan |
| author_sort | Muller, Christiaan |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Thesis (PhD)--Stellenbosch University, 2024. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/130484 |
| institution | Stellenbosch University (South Africa) |
| language | en_ZA en_ZA |
| last_indexed | 2026-06-10T12:43:27.297Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2024 |
| publishDateRange | 2024 |
| publishDateSort | 2024 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/130484 Map point selection for hardware constrained visual simultaneous localisation and mapping. Muller, Christiaan Van Daalen, Corne E. Stellenbosch University. Faculty of Engineering. Dept. of Electrical and Electronic Engineering. Robotics -- Control systems Intelligent control systems Simultaneous localisation and mapping UCTD Thesis (PhD)--Stellenbosch University, 2024. ENGLISH ABSTRACT: Simultaneous localisation and mapping (SLAM) plays a vital role in autonomous robotics. Robotic platforms are often resource constrained, and this limitation necessitates resource efficient SLAM implementations. While sparse visual SLAM algorithms offer reasonable accuracy for modest hardware requirements, these more scalable approaches still face limitations when applied to large-scale and long-term scenarios. Existing SLAM approaches require manual adjustment to maintain a compact map of the environment. This dissertation presents an optimisation-based approach to select a good subset of map points for visual SLAM, which discards redundant map points in an automated fashion. Novel information-theoretic approaches are developed as solutions for this optimisation problem. Existing coverage-based approaches for related robotics problems are also adapted to selecting map points for visual SLAM. The first of the novel information-theoretic approaches developed in this dissertation allows for accurate SLAM estimation by considering the full SLAM estimation problem, which allowed the reduction of the map points by more than 60%. This is in contrast to coverage-based approaches, which do not allow for accurate SLAM estimation. This first information-theoretic approach is too computationally expensive to use online and motivates the development of computationally inexpensive alternatives. To this end, this dissertation also proposes two additional information-theoretic map point selection techniques that approximate the SLAM estimation problem. These approximate approaches allow for similar trajectory accuracy to the original approach while reducing computation times from hours to often less than a second. These approximate approaches enable the practical online application of these techniques to reduce maps, while minimising the impact on SLAM trajectory accuracy. Lastly, this dissertation also presents a combined approach that improves on the approximate information-theoretic approach by including a coverage-based model. This combined approach outperforms all alternative map point selection approaches developed in this dissertation. Map point selection approaches are first evaluated offline using a modified version of an existing visual SLAM algorithm (ORB-SLAM 2). Different map point selection approaches are evaluated on practical data from both the EuRoC and KITTI datasets. A follow-up experiment also demonstrates the application of map point selection approaches online. AFRIKAANSE OPSOMMING: Gelyktydige lokalisering en kartering (GLEK) speel ’n belangrike rol in outonome robotika. Robotiese platforms is dikwels hulpbron beperk, en hierdie beperking noodsaak hulpbron doeltreffende GLEK-implementerings. Terwyl yl visuele GLEK-algoritmes redelike akkuraatheid bied vir beskeie hardeware vereistes, staar hierdie meer skaalbare benaderings steeds beperkings in die gesig wanneer dit toegepas word op grootskaalse en langtermynscenario’s. Bestaande GLEK-benaderings vereis handaanpassing om ’n kompakte kaart van die omgewing te handhaaf. Hierdie proefskrif bied ’n optimeringgebaseerde benadering om ’n goeie subversameling van kaartpunte vir visuele GLEK te kies, wat oortollige kaartpunte op ’n outomatiese wyse weggooi. Dit bied nuwe inligting teoretiese benaderings om oplossings vir hierdie optimerings probleem te ontwikkel. Bestaande dekkinggebaseerde benaderings vir verwante robotika probleme is ook aangepas om kaartpunte vir visuele GLEK te kies. Die eerste van die nuwe inligting teoritiese benaderings wat in hierdie proefskrif ontwikkel is, maak voorsiening vir akkurate GLEK-skatting deur die volle GLEK-skattings probleem te oorweeg, wat die vermindering van die kaart punte met meer as 60% moontlik gemaak het. Dit is in teenstelling met dekkinggebaseerde benadering, wat nie voorsiening maak vir akkurate GLEK-beraming nie. Hierdie eerste inligting teoretiese benadering is rekenkundig te duur om aanlyn te gebruik en motiveer die ontwikkeling van rekenkundig goedkoop alternatiewe. Vir hierdie doel stel hierdie proefskrif ook twee bykomende inligting teoretiese kaartpuntseleksietegnieke voor wat die GLEK-beraming probleem benader. Hierdie benaderings maak voorsiening vir soortgelyke trajek akkuraatheid as die oorspronklike benadering, terwyl die berekening tye van ure tot dikwels minder as ’n sekonde verminder word. Gevolglik maak hierdie benaderings die praktiese aanlyn toepassing van hierdie tegnieke moontlik. Laastens bied hierdie proefskrif ook ’n gekombineerde benadering aan wat die vereenvoudige inligting teoretiese benadering verbeter deur ’n dekkinggebaseerde model in te sluit. Hierdie gekombineerde benadering presteer beter as alle alternatiewe kaartpuntseleksie benaderings wat in hierdie proefskrif ontwikkel is. Kaartpuntseleksie benaderings word eers aflyn ge¨evalueer deur gebruik te maak van ’n gewysigde weergawe van ’n bestaande visuele GLEK-algoritme (ORB-SLAM 2). Kaartpuntseleksie benaderings word op praktiese data van beide die EuRoC- en KITTI-datastelle ge¨evalueer. Daarna word die tegnieke aanlyn gedemonsteer in ’n opvolgeksperiment. Doctorate 2024-02-29T08:00:27Z 2024-04-26T19:17:27Z 2024-02-29T08:00:27Z 2024-04-26T19:17:27Z 2024-03 Thesis https://scholar.sun.ac.za/handle/10019.1/130484 en_ZA en_ZA Stellenbosch University ix, 137 pages : illustrations application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Robotics -- Control systems Intelligent control systems Simultaneous localisation and mapping UCTD Muller, Christiaan Map point selection for hardware constrained visual simultaneous localisation and mapping. |
| title | Map point selection for hardware constrained visual simultaneous localisation and mapping. |
| title_full | Map point selection for hardware constrained visual simultaneous localisation and mapping. |
| title_fullStr | Map point selection for hardware constrained visual simultaneous localisation and mapping. |
| title_full_unstemmed | Map point selection for hardware constrained visual simultaneous localisation and mapping. |
| title_short | Map point selection for hardware constrained visual simultaneous localisation and mapping. |
| title_sort | map point selection for hardware constrained visual simultaneous localisation and mapping |
| topic | Robotics -- Control systems Intelligent control systems Simultaneous localisation and mapping UCTD |
| url | https://scholar.sun.ac.za/handle/10019.1/130484 |
| work_keys_str_mv | AT mullerchristiaan mappointselectionforhardwareconstrainedvisualsimultaneouslocalisationandmapping |