Details of Award
NERC Reference : NE/V019759/1
Using low-cost, wearable, sensors to detect driver state: Towards development of new algorithms for real-time detection of driver fatigue
Grant Award
- Principal Investigator:
- Professor N Merat, University of Leeds, Institute for Transport Studies
- Grant held at:
- University of Leeds, Institute for Transport Studies
- Science Area:
- None
- Overall Classification:
- Unknown
- ENRIs:
- None
- Science Topics:
- Intelligent Transport Systems
- Transport Ops & Management
- Transport Safety
- Road Safety
- Cognitive Psychology
- Experimental Psychology
- Technology and method dev
- Abstract:
- To date, there has been limited research exploring the feasibility, accuracy and power of sensor fusion of different low-cost physiological sensors, such as those that measure skin conductance, heart rate variability and eye tracking, to estimate driver state. The accurate, real-time, estimation of driver states such as drowsiness can provide support and warning to drivers, prompting them of their impairment, encouraging them to take brakes, and ultimately improving road safety. To increase the availability of such devices in the future, there is need to develop more affordable, less cumbersome devices, compared to the research grade alternatives used in research laboratories. The aim of this project is to collect a range of drivers' physiological data, using low-cost consumer grade sensors, applying sensor fusion to create suitable machine learning algorithms that can accurately identify drivers' drowsy state. My project aims to answer the following two main research question: i. Can driver drowsiness be accurately detected using low-cost physiological sensors, ii. Can sensor fusion and machine learning algorithms be used to provide real-time feedback to drowsy drivers? The objectives of this project are to: Assist and co-ordinate in design of the simulator experiment at UoT, to measure drowsiness during driving using low-cost physiological sensors, prior to arrival in Canada. Assist in data collection, on site, and lead on the pre-processing of physiological signals. Co-develop suitable machine learning algorithms that use sensor fusion, to accurately detect driver state, such as drowsiness/fatigue. My project is part of larger research project titled - "Estimating drowsiness and high cognitive load in drivers: A driving simulation experiment", being conducted at the UoT. The project includes collection of data from various vehicle-based sensors, and physiological sensors such as electroencephalography (EEG), electrocardiography (ECG), Photoplethysmogram (PPG), electrodermal activity (EDA), and eye-tracking, for identifying driver states of drowsiness and high cognitive load. My expertise in data reduction and processing of physiological metrics during my PhD at Leeds is considered attractive by UoT, hence the rationale for our proposed collaboration.
- NERC Reference:
- NE/V019759/1
- Grant Stage:
- Completed
- Scheme:
- NC&C NR1
- Grant Status:
- Closed
- Programme:
- Globalink Placement
This grant award has a total value of £10,585
FDAB - Financial Details (Award breakdown by headings)
| Exception - Other Costs |
|---|
| £10,585 |
If you need further help, please read the user guide.