About the Project
The project would see the combination of HCI/UX, driving simulation software+hardware and data analysis for users of vehicles (fossil fuelled or electric powered).
Through use of a set of race tracks as well as open-world roads, available within the simulation software currently installed, it would be possible to reproduce real-world scenarios that vehicles and their drivers go through regularly, and from that, capture the driver actions as they react to set conditions. This can include weather simulation, such as rain.
This can be completed in a risk free, but realistic fashion.
The data collected from the test drivers would be valuable for aiding analysis of classification of the key elements that drivers should exhibit to be deemed to have passed any automotive examination test of their ability to control a vehicle and to grade their level of skill.
Models of human driving behaviours would be constructed and classified with those that are desirable and safe being recommended.
There would be a range of approaches for detecting and classifying human activity during driving a vehicle using a variety of sensors and telemetry available from the simulator.
It will then possible to incorporate the characteristics identified and build a modelled world simulation that has typical roads, conditions and AI generated drivers.
The aim of this research is to establish an approach to identify, classify and recommend vehicle usage characteristics that could be used to help examiners of vehicle skill level to ensure most reliably that a driver meets required standards.
The core objectives will be to:
(1) identify the human driving style as a set of classifiable characteristics;
(2) establish how best to recognise driver and skill level;
(3) to explore how best to gain and visualise the insights from the vast level of telemetry data;
(4) reproduce and build pseudo-drivers with modelled behaviour and then be able to use this to examine driver behaviour as they interact with the system.
These objectives will necessitate making use of machine learning techniques to develop models that can then be used for driver classification and the characteristics to control a vehicle and then to add the AI drivers to the system.
This proposal fits with the University’s strategic theme of sustainability, relating to digital futures, environmental sustainability and computing. It is also closely aligned with both the Pervasive Computing Research Group and AI Research Centre, focusing upon research within the areas of Human Computer Interaction and Data Analytics / AI. The project benefits from access to a range of existing pervasive and wearable sensing technologies, notably the SimLab P1X aluminium cockpit with integrated computer and car simulation software, which can be used for data collection and model testing. The supervisory team has extensive expertise and experience in both the theory surrounding the work and its application.
Booth, FG, R Bond, R, D Mulvenna, M, Cleland, B, McGlade, K, Rankin, D, Wallace, J & Black, M 2021, 'Discovering and comparing types of general practitioner practices using geolocational features and prescribing behaviours by means of K-means clustering: A Comparison of Prescribing Behaviours Between Practice Types', Scientific Reports, vol. 11, no. 1, 18289, pp. 1-15. https://doi.org/10.1038/s41598-021-97716-3
Pita-Costa, J, Rei, L, Stopar, L, Fuart, F, Grobelnik, M, Mladenić, D, Novalija, I, Staines, A, Pääkkönen, J, Konttila, J, Bidaurrazaga, J, Belar, O, Henderson, C, Epelde, G, Arrúe Gabaráin, M, Carlin, P & Wallace, JG 2021, 'NewsMeSH: a new classifier designed to annotate health news with MeSH headings', Artificial Intelligence in Medicine, vol. 114. https://doi.org/10.1016/j.artmed.2021.102053
Rankin, D, Black, M, Flanagan, B, Hughes, C, Moore, A, Hoey, L, Wallace, J, Gill, C, Carlin, P, Molloy, A, Cunningham, C & McNulty, H 2020, 'Identifying Key Predictors of Cognitive Dysfunction in Older People Using Supervised Machine Learning Techniques: Observational Study', JMIR Medical Informatics, vol. 8, no. 9, e20995, pp. 1-23. https://doi.org/10.2196/20995
Johnston, V, Black, M & Wallace, JG 2020, A Holistic UX Methodological Framework for Measuring the Aspects of How Dynamic, Adaptive and Intelligent a Software Solution is and Make Recommendations for Improvement. in Collaborative European Research Conference (CERC 2020).
L Yang, S McClean, M Donnelly, K Burke, K Khan, (2022), A multi-components approach to monitoring process structure and customer behaviour concept drift, Expert Systems with Applications 210, 118533.
L Yang, S McClean, M Donnelly, K Khan, K Burke (2020), Analysing Business Process Anomalies Using Discrete-time Markov chains, IEEE 22nd International Conference on High Performance Computing and Communications; IEEE 18th International Conference on Smart City; IEEE 6th International Conference on Data Science and Systems (HPCC/SmartCity/DSS)
I Cleland, MP Donnelly, CD Nugent, J Hallberg, M Espinilla, (2018) Collection of a diverse, realistic and annotated dataset for wearable activity recognition,
2018 IEEE International Conference on Pervasive Computing and Communications
M Espinilla, J Medina, A Salguero, N Irvine, M Donnelly, I Cleland, and Chris Nugent, (2018) Human activity recognition from the acceleration data of a wearable device. which features are more relevant by activities?, Multidisciplinary Digital Publishing Institute Proceedings 2 (19), 1242
M. Jensen, J. Wagner and K. Alexander, 2011, October, Analysis of in-vehicle driver behaviour data for improved safety, International Journal of Vehicle Safety, Vol. 5, No. 3, pp 197-212
Van Laerhoven, K., Aidoo, K.A. and Lowette, S., 2001, October. Real-time analysis of data from many sensors with neural networks. In Proceedings fifth international symposium on wearable computers (pp. 115-122). IEEE.
Yuan-Lin Chen and Wei-Jen Lee, 2011, Safety distance warning system with a novel algorithm for vehicle safety braking distance calculating, International Journal of Vehicle Safety, Vol. 5, No. 3, pp 213-231
A.Hamish JamsonNatashaMerat, 2005, March, Surrogate in-vehicle information systems and driver behaviour: Effects of visual and cognitive load in simulated rural driving, Transportation Research Part F: Traffic Psychology and Behaviour, Volume 8, Issue 2, pp 79-96.
Charalampos M., Erotokritos X., Liana C. 2017, July, Simulation of electric vehicle driver behaviour in road transport and electric power networks, Transportation Research Part C: Emerging Technologies, Volume 80, pp. 239-256
S.HelmanN.Reed1, 2015, February, Validation of the driver behaviour questionnaire using behavioural data from an instrumented vehicle and high-fidelity driving simulator, Accident Analysis & Prevention, Volume 75, pp. 245-251
Ahmed Al-Hussein W, Mat Kiah ML, Lip Yee P, Zaidan BB. 2021. A systematic review on sensor-based driver behaviour studies: coherent taxonomy, motivations, challenges, recommendations, substantial analysis and future directions. PeerJ Computer Science 7:e632