Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.486511
Title: Remote access to a prototyping laboratory
Author: Lathwell, Stephen
ISNI:       0000 0001 3605 4164
Awarding Body: Open University
Current Institution: Open University
Date of Award: 2007
Availability of Full Text:
Access through EThOS:
Abstract:
There is a growing global demand for continuing adult higher education particularly in science and engineering subjects. New technologies are emerging which would enable the development of a Remote Access Laboratory for rapid prototyping of Artificial Intelligence, as a learning environment for mechatronic engineering, in which high precision electromechanical devices are designed to exhibit autonomous behaviour. Secondary research investigated the learning theories for a Remote Access Laboratory, and the current practices for distance learning, involving groupware in shared activity 'collaboratories'. Having determined that the laboratory would need a multi-user interactive environment architecture, with the requirement for adaptability to rapid developments, a distributed software architecture was selected. The laboratory design was subsequently argued to be best served by IntelligentAgents in a Multi-Agent system. The aims ofthe research were to establish the viability of a Remote Access Laboratory , for mechatronic experimentation, and to evaluate the technologies required to implement such a laboratory environment for rapid prototyping. These were achieved by developing a novel user interface, based on a multi-functional screen layout, and a graphical specification facility to provide robotic navigation that is intuitive to use and does not require text-based programming. The research investigated the prototyping ofrobotic behaviour, which used Programming by Demonstration as an innovative technique to prototype robot navigation. The method of designjng behaviours met an anticipated need to allow the robot to interact with an environment, to achieve goals under conditions ofuncertainty, while requiring a level of abstraction in the behaviour design. The interface structured a composite ofthe designed behaviours into prototype Artificial Intelligence using' a hierarchical behaviour architecture, which complied with the principles ofObject Orientated programming. This was subsequently anew and original programming method to facilitate rapid prototyping ofArtificial Intelligence design and structuring. Experimentation involved 20 participants attempting to accomplish a series of tasks which involved using the prototyped interface and an existing text-based robot programming system. The participants were profiled by their formal qualifications, knowledge and experience. The experimental data obtained were used to establish a comparative measure of the prototype interface success compared with an existing distance-learning, home experiment kit, in the form ofa small controllable model vehicle. The data obtained provided strong evidence to support the hypothesis that a Programming by Demonstration based system for rapid prototyping is more flexible and easier to use than a previously existing distance learning text-based system. The Programming by Demonstration system showed great promise, being quicker for prototyping, and more intuitive. The learning interface design pioneered new techniques and technologies for rapid prototyping of Artificial Intelligence in a Mechatronics Remote Access Laboratory.
Supervisor: Not available Sponsor: Not available
Qualification Name: Open University, 2007 Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.486511  DOI: Not available
Share: