A new tool has been developed called RoadSafe designed to alert fleet drivers to high-risk highways.
The tool, created by a UK government-backed consortium that’s co-funded by Innovate UK, involved a 20-month project by organizations including Ford, Loughborough University, Oxfordshire County Council, and AI sensor specialists Vivacity Labs. It also has the support of Transport for London.
Natalie Middleton writes in her Fleet News article of September, Ford ‘RoadSafe’ dashboard to alert fleet drivers to high-risk roads that the digital tool could alert fleet drivers to accident blackspots and significantly reduce road accidents. However, a spokesperson for Ford says its more about helping local authorities to manage busy roads to reduce congestion by re-directing traffic away from busy areas or closed roads, as well as to ensure that accidents are minimized.
Middleton reports that with the UK consortium, Ford is working “to process anonymized data from sources including connected vehicles, roadside sensors and accident reports via a smart algorithm and pinpoint where there is a higher chance of traffic incidents occurring”.
Jon Scott, project lead, City Insights, Ford Mobility, Europe, said: “There are areas in every city where the chance of an incident is higher, whether it’s due to a poorly placed sign, an unrepaired pothole or junctions built to accommodate far less traffic than we have today. Now, Ford can pinpoint the areas of concern, so drivers could be made more aware of them, and authorities can address them.”
Ford says the research began with an analysis of Greater London “to highlight road safety hotspots and to identify the potential causes and safety mitigations”. The research then expanded over a period of 15 months to Oxfordshire. The automaker says it involved more than 200 passenger and commercial vehicles voluntarily connected in London and Oxfordshire. The data that emerged from the project, and particularly from the road-side artificial intelligence (AI) sensors, enabled the teams to develop a road segment risk heat map to identify stretches of road that are of particular concern.
Joseph Salem, partner head of transport for the Middle East and India, at consultancy firm Arthur D. Little also believes that the digitalization of infrastructure is coming hand-in-hand to enable the full potential of connected vehicles.
He explains: “Most manufacturers are looking to leverage the intelligence of the infrastructure to push value added services to the customers. There is a wish to improve the overall customer experience, but mostly they want to improve safety. This trend is expected to further accelerate as indicated by the recent wave of venture capital investments to push for the autonomous and connected mobility systems between the vehicle and infrastructure.”
He says two-thirds of the new vehicles sales involve connected vehicles, and that there has been over $2M invested in infrastructure and digitalization in the first half 2021. “This is the main strand for mobility,” he suggests.
Improving road safety
Steven Reed, research associate at Loughborough University’s Transport Safety Research Centre, which is part of the Loughborough Design School, says the focus of his own team within the RoadSafe project is very much on improving road safety. Yet he admits that most people don’t expect to have a collision when they leave their house or place of work. Most people view the chances of having an accident as being very small, and this misleads drivers to think that the road is a safe place.
“However underlying risk exists across every part of the UK road network and work undertaken for the Road Safe tool at Loughborough University has been to identify these inherent risks,” he explains before adding that some of the risks are relatively obvious to the typical road user. For example, the effect of drink and drug-driving has on driving ability and safety. Yet he finds that there are other risks that are more benign such as the effect of junction density, and effect it has on crash risk, or the effects of different traffic compositions in terms of vehicle types and pedestrian activity.
He adds: “Collisions based on these risk factors don’t always occur as these indicators often remain latent within the traffic system until the right conditions are met and from an individual driving perspective remain mostly invisible reinforcing the notion that the roads are otherwise safe. A raft of these established indicators exist within road safety risk literature which can be used to measure or predict road safety performance. From a road safety perspective, this is primarily the aim of the tool, to highlight factors such as these within a simple and easy to digest risk rating.”
Google and Waze, he says, predict and detect traffic holdups. RoadSafe does much the same to highlight areas that considered to be particularly dangerous. It can be used to inform drivers, including fleet drivers, of the potential risk of being involved in a collision, or it can tell them about the risk of being caught up in traffic as a result of other vehicles being involved in a collision.
Reed and his team recognize that not all accidents hard drivers or other road users. However, talking from Loughborough University’s perspective the benefit of working with Ford, he says: “The benefit of working with Ford is that the tool can access valuable data from Ford vehicles to highlight areas where road users struggle with specific road design or regularly undertake unsafe driving manoeuvres.”
“Using Loughborough University’s extensive experience from undertaking real-world driving studies a series of ‘triggers’ have been refined and defined to identify potentially unsafe acts based on vehicle dynamics and other vehicle data. This data is used to monitor the live performance of road networks and specific junctions, the outcome of this process generates a potential to warn drivers of upcoming hazards, allowing a driver to take more care over their driving or in some cases avoid the area entirely to reduce the chance of delays.”
Hurdles of complexity
He believes that the use of vehicle data is very “powerful for a global understanding of a road network.” However, there is a slight problem. In complex areas that have diverse traffic flows, he reveals that the vehicle “may not always yield the required detail on its own.” To overcome this hurdle, he advises that an accident data-driven approach must be used. This can nevertheless be time-consuming, and he suggests it can be ethically questionable “as many road users will need to be exposed to the predicted risk before a long enough accident history is evident”.
“To accelerate this process and minimize the risk a complimentary approach of monitoring traffic conflicts alongside the vehicle data has been developed,” he says before commenting that this process “targets established risk measures that can predict the likelihood and subsequent severity of collisions occurring.”
The process involves deploying and developing smart sensors by Vivacity Labs. They are placed in selected locations of concern for the monitoring of traffic behavior and to calculate the road safety measures defined by the university. The sensors use a camera and a local processor to extract anonymous data on how roads users are using the space within the sensors’ field of view. It also includes data on pedestrians and cyclists, speed, location, and proximity to other road users. Once processed into the required metrics, this data can help to create “a detailed understanding of how specific road segments are performing even before collisions occur”.
Reed finds that this approach is particularly powerful when it comes to monitoring the effectiveness of new traffic schemes, modal change, and new technologies such as connected and autonomous vehicles (CAVs), or the uptick of cycling traffic. It is particularly useful when no historic traffic or accident data is unavailable.
Ford therefore concludes: “In the future, such technology could also benefit passengers riding in autonomous vehicles. Combining the on-board sensors of the vehicle with a digital tool could help them anticipate hazardous situations even earlier and therefore adapt their operation accordingly.” Digital dashboards therefore have a wider use beyond fleet management as road safety is their primary concern. Reducing congestion also saves time, money, and reduced vehicle tailpipe pollution.
by Graham Jarvis