Two years in, East Riding of Yorkshire Council’s innovative road safety and decarbonisation Live Labs 2 pilots are progressing at a pace – and the results could yet change street lighting as we know it profoundly, writes Karl Rourke.
For those who are not aware, Adept Live Labs 2 is a DfT-funded £30m research and innovation project, over three years, that is looking at profiling the embedded carbon in our highways assets and our highways operations from fence to fence.
Our programme at East Riding of Yorkshire is focusing on those assets that provide road users with the visual information that they need to navigate the road in a safe manner. So, specifically street lighting, signage, and markings.
We’re focusing on A roads and bypass routes. Crucially, we are looking at the areas of these roads that are specifically required to be lit. So, conflict zones, roundabouts, crossroads and priority junctions.
We’re asking the question: do they actually need to be lit? Can we do them in a different way? And can we reduce the carbon and financial burden of these locations? We’ve got 22 test sites on our two busiest routes in East Riding and seven on sites in other areas of the UK.
We’re testing an alternative design where, instead of street lighting, we’re using solar-powered illuminated studs for lane delineation; improved signage; and increased reflectivity of white lines. The aim is to give road users the visual prompts they need to get around safely.
In areas where there is pedestrian usage, we are providing lighting only for the pedestrians, in other words designed only to light the width of the footway.
You can imagine the responses we get when we say we’re going to remove lighting off conflict zones. ‘It will be carnage’, is what is generally written all over people’s faces. So, it is important to emphasise that ‘user safety’ is a critical outcome of the success of this project alongside carbon reduction.
AI ANALYSIS OF COLLISION DATA
We’re looking at road safety in a way never done before on lit or non-lit roads. Underpinning it all are the road safety risk assessments we’ve undertaken on all test sites, including in line with GG104, and the road safety audits that have been completed on all sites.
But all of this has a very specific, non-standard Live Labs-focused brief. Where lighting is placed as an absolute last resort.
A standard risk assessment will review collision data for the previous three years. But, in order to be fully confident with our approach, we’ve reviewed collision data going back 10 years. And it has thrown up some very interesting results.
What we’ve found is, firstly, the collision record during the hours of darkness is 20% for the crossroads and 11% for the roundabout. Across all sites combined, it still only comes at an average of 20% of collisions happening at night.
You might think, ‘well that’s the lighting doing its job’. Not necessarily. Because we then bring in another key statistic – the most common form of collision. For the crossroads, it is a failure to give way. And for the roundabout, it is a rear-end shunt. Both of these occur no matter the time of day. And both of them are directly attributable to driver error. No amount of engineering or lighting is going to stop a driver not paying attention.
What our statistics do pull out is that traffic volumes and driver behaviour have more of a bearing on collision records than lighting. Our assets work when traffic volumes are generally light – less traffic, fewer collisions.
Now, I accept you might be thinking this is quite a big leap to take. Let me, in answer, introduce you to our road user behaviour AI system. I have affectionately nicknamed it ‘the other half’ because it has been very expensive, absolutely exhausting, and it sees everything you do, even in the dark! Or maybe that’s just my other half!
Most ILP members, I am sure, will have some awareness of visual-based AI systems for traffic monitoring. Our system, I believe, takes things to a whole new level. Like I said, it can ‘see’ in the dark. We’ve used thermal imaging, and we’ve adjusted and retrained an existing AI system to accommodate this change of image. This type of system is a key new innovation for the project and the first of its type in the marketplace.
It works – it provides absolute clarity of data and information, through all vehicle types and, in fact, including pedestrians, cyclists and motorcyclists. It gives us average speeds, and individual speeds, for each individual carriageway and each user type. It plots reaction times, it gives us a safety index scoring, it identifies a list specifically relating to individual conflicts. And it plots the conflict point on the road, if the drivers are not reacting. It even gives movement plots for pedestrians, which will be very interesting in the location that we will be switching over to only-footway lighting.
The system gives us a full spectrum of driver indicators, down to absolute, granular details. Not only will it be used for driver behaviour monitoring for the next five years, it can also be used potentially for accident recovery and even in criminal investigation. Because alongside the thermal imaging cameras, we have optical cameras built into the same housing.
TEST SITES GOING LIVE
This data is the cornerstone of our Live Labs project. It will prove that our approach is safe. We’ve taken a full week’s worth of video – from a week in mid-September last year – we’ve processed that against a ‘business as usual’ baseline. So normal street lighting, normal daylight, normal lighting conditions.
Over the next few weeks we will move to a real-time processing version as we begin to make our test sites live on a large scale. Plus, we have control sites that will be monitored alongside the test sites, and they will be remaining lit to provide a constant comparison as we move through the seasonal variances.
Our project already attracted a lot of media attention during the autumn, in particular for our pedestrian-only lighting. Early in the project, as test sites were identified, one thing we noticed is that some of the sites had the audacity to have well-used footways alongside. I mean, how dare they!
So, barring rolling out head torches for all residents in East Riding, it was clear we needed to be providing lighting for non-vehicular users. We of course do not want to be mentally shutting people in their own homes after dark. Quite the opposite: we want people to be getting out and about and enjoying the night-time space.
LOOKING TO THE FUTURE
The next phase of this element of work, which we are well on the way with, is that we are creating a carbon cost-evaluation tool with Leeds University. This will incorporate the lifespans of the new materials, both either street lighting or highway, and it will look at embedded whole-life carbon and operational whole-life carbon costs. It will be available for trial use within the project from this summer.
It is, essentially, an assessment tool not a baselining tool. It will be a tool for designers and decision-makers to decide on whether to go with a standard street lighting approach or to go with the new Live Labs methodology. It will also incorporate solar-powered lighting; as we speak we are beginning a full whole-life carbon appraisal of solar-powered versus mains-powered lighting. Set in that grid decarbonisation context.
The outcomes of that, I think, are going to be very, very interesting. We need to understand that saving energy is not our only contribution to net zero as an industry. Things are changing, but they are changing slowly. The message is getting through, but it must get through faster.
Finally, what’s next? The main focus over the next few months is to get the pedestrian lighting test-beds installed, and get the partner test-beds installed. We will have a heavy focus on the biodiversity engagement and the carbon assessment tool. I have absolutely no doubt there will be a significant comms flurry when the pedestrian lighting goes in.
Then, as we get to the autumn and the winter, we will be preparing our final statements and findings. We will be doing our AI data evaluation. I will look forward to telling you next year what we have found and what our conclusions are.
Karl Rourke is service manager, Street Lighting, Traffic Signals and CCTV, at East Riding of Yorkshire Council
This is an abridged version of the article that appears in the April edition of Lighting Journal. It was also based on a talk Karl gave to the Lighting Live – Local Authority ILP event in Daventry in February. To read the full article, simply click on the page-turner to your right.
