Active strategies for coordination of solitary robots

Masakuna, Jordan Felicien (2020-12)

Thesis (PhD)--Stellenbosch University, 2020.

Thesis

ENGLISH ABSTRACT: This thesis considers the problem of search of an unknown environment by multiple solitary robots: self-interested robots without prior knowledge about each other, and with restricted perception and communication capacity. When solitary robots accidentally interact with each other, they can leverage each other’s information to work more effectively. In this thesis, we consider three problems related to the treatment of solitary robots: coordination, construction of a view of the network formed when robots interact, and classifier fusion. Coordination is the key focus for search and rescue. The other two problems are related areas inspired by the problems we encountered while developing our coordination method. We propose a coordination strategy based on cellular decomposition of the search environment, which provides sustainable performance when a known available search time (bound) is insufficient to cover the entire search environment. A sustainable performance is achieved when robots that know about each other explore non-overlapping regions. For network construction, we propose modifications to a scalable decentralised method for constructing a model of network topology which reduces the number of messages exchanged between interacting nodes. The method has wider potential application than mobile robotics. For classifier fusion, we propose an iterative method where outputs of classifiers are combined without using any further information about the behaviour of the individual classifiers. Our approaches for each of these problems are compared to state-of-the-art methods.

AFRIKAANSE OPSOMMING: Hierdie tesis beskou die probleem van soektog in ’n onbekende omgewing deur ’n aantal alleenstaande robotte: selfbelangstellende robotte sonder voorafgaande kennis van mekaar, en met beperkte persepsie- en kommunikasievermoëns. Wanneer alleenstaande robotte toevallig mekaar raakloop, kan hulle met mekaar inligting uitruil om meer effektief te werk. Hierdie tesis beskou drie probleme wat verband hou met die hantering van alleenstaande robotte: konstruksie van ’n blik van die netwerk gevorm deur interaksie tussen robotte, koördinasie en klassifiseerdersamesmelting. Koördinasie is die hoof fokuspunt vir soek en redding. Die ander twee probleme is uit verwante areas, gemotiveer deur uitdagings wat ons ervaar het tydens die ontwikkeling van ons koördineringsmetode. Ons stel ’n skaleerbare desentraliseerde metode voor om ’n model van netwerktopologie te bou wat minder boodskappe tussen wisselwerkende nodusse hoet te verruil. Die metode het wyer potensiële toepassings as mobiele robotika. Vir koördinasie, stel ons ’n strategie voor gebaseer op sellulêre ontbinding van die soekomgewing, wat volhoubare prestasie toon wanneer ’n bekende soektyd onvoldoende is om die hele soekomgewing te dek. Vir klassifiseerdersamesmelting, stel ons ’n iteratiewe metode voor, waar klassifiseerders se voorspellings gekombineer word sonder om enige verdere inligting oor die gedrag van die individuele klassifiseerders te gebruik. Ons benaderings vir elkeen van hierdie probleme word vergelyk met stand-van-die-kuns metodes.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/109111
This item appears in the following collections: