A virtual reality-based context simulator for evaluating the effect of uncertainty on Smart Building Applications
Citation:
Kris McGlinn, 'A virtual reality-based context simulator for evaluating the effect of uncertainty on Smart Building Applications', [thesis], Trinity College (Dublin, Ireland). School of Computer Science & Statistics, 2013, pp 137Download Item:
Abstract:
Smart Building Applications (SBAs) adapt their behaviour in response to context information
generated by sensors present in the smart building. SBA developers must however account for
uncertainty in the accuracy of context information resulting from imperfection in the sensing
technology and potentially complex interactions between the sensed phenomenon, the sensor and
the physical environment. To address this challenge, Context Simulators are increasingly employed.
These generate simulated context information based on an internal model of the building, it sensors
and its occupants and their activities. Some simulators employ Virtual Reality (VR) environments, to
more accurately capture the likely behaviour of users in a simulated smart building, and thereby
model and simulate their interaction with sensors and, via the resulting context information, with
SBAs. This type of evaluation can result in better designed SBAs by prototyping them in simulated
conditions before being deployed into a real physical environment. A number of key challenges
remain to be addressed however, before such a simulation-based SBA prototyping approach can
become widely used. Firstly, any context simulator must be flexible in its approach to simulating
context, so that it supports simulating different mixes of SBAs, context sources and smart buildings,
using a selection of methods most suitable to the evaluation task at hand. Secondly, the context
simulator must support simulating uncertainty in context. Lastly, the process of setting up a
simulation should be efficient in use, satisfying to the developer and also, be effective in correctly
evaluating the SBA behaviour.
This thesis presents the SimCon model which addresses these challenges by supporting simulation of
context, in particular location context, with configurable levels of uncertainty for evaluating SBAs
using visualisation. A comparative analysis of the state of the art is used to demonstrate the relative
flexibility of the model. The SimCon model has been implemented in a simulation configuration and
visualisation tool set which has undergone a number of evaluations with SBA developers to
determine and improve its level of usability. Integration with existing building industry standards has
also been developed, to increase the leverage of such models and thereby the efficiency of the
simulation configuration process.
Author: McGlinn, Kris
Advisor:
Lewis, DavePublisher:
Trinity College (Dublin, Ireland). School of Computer Science & StatisticsNote:
TARA (Trinity's Access to Research Archive) has a robust takedown policy. Please contact us if you have any concerns: rssadmin@tcd.ieType of material:
thesisAvailability:
Full text availableMetadata
Show full item recordLicences: