HAROLD:

HAzards, ROad Lighting and Driving

Start Date: March 2019
Staff: Steve Fotios, Chris Cheal, Yichong Mao, Chloe Robbins
Funding: EPSRC EP/S004009/1

In progresss

The frequency and severity of road collisions are reduced if drivers are able to detect and recognize potential hazards in sufficient time to take evasive action such as braking and steering. An improvement in detection time measurable in hundreds of milliseconds could substantially increase the probability that a crash can be avoided. 

After dark, visual functions such as reaction time are significantly reduced, and road lighting is installed as a countermeasure to this visual impairment. Road lighting is of particular importance for revealing hazards beyond the reach of vehicle headlights such as pedestrians emerging from the side. 

This project investigates the way in which lighting can be used to enhance safety on main roads. Specifically, we seek lighting that increases the chance of drivers seeing a hazard and reduces the time taken to see the hazard. These hazards include other vehicles, stationary objects and pedestrians. 

There are two problems with the current situation. First, while British and European standards provide guidance on road lighting, the empirical basis of the recommended lighting is not clear. Therefore, we do not know whether they recommend optimal conditions. Second, we suspect there is a better way for enhancing the detection of pedestrians when they are otherwise unexpected, which is frequently the case for pedestrians on main roads. This is that pedestrians should use a pulsing or flashing LED band, worn on the wrist or ankle to take advantage of bio-motion.  An LED band could provide a low-cost counter-measure to reduce the risk of accident. 

To investigate these proposals we will first carry out experiments to find out how the detection of hazards including pedestrians is affected by changes in lighting, using variations in the intensity and spectrum (colour) of lighting. Whilst drivers should be continuously scanning for potential hazards, there are many distractions – listening to music, speaking to passengers and looking at maps or digital navigation devices. These distractions reduce our ability to detect hazards. We will therefore also investigate how hazard detection is affected by distraction and whether optimal lighting can mitigate the distraction decrement. This research is of particular benefit to elderly drivers; the elderly tend to have poorer vision and, overall, they perform worse than younger people when driving with distractions

From these data we will identify the changes in lighting conditions likely to improve safety. These recommendations will be validated by manipulating lighting conditions within a high fidelity driving simulator at the University of Leeds. The simulator places the test participant in a more realistic setting while still maintaining control on road situation and ensuring participant safety. To facilitate implementation of results we will work to ensure the guidance and standards documents used by lighting designers are revised to include the proposed criteria.

Ongoing work

Progress on HAROLD was severely impacted by the Covid pandemic: we were about to start experimental work in March 2020 when building closure and social distancing rules were introduced. During that period we instead focussed on analyses of secondary data - road collisions, crime and traffic flow - as there are issues which also affect the optimisation of road lighting criteria. A one-year no-cost extension was approved. An experiment was conducted jointly with a PhD student in the LightCAP project - the results are currently being analysed.

Publications