Increasing safety on roads is high on many agendas these days – from policymakers and campaign groups to everyday road users and commercial fleet operators.

But rather than waiting for legislative change, fleet operators and managers can seize AI-enabled technology now – and start improving work-related road safety today. In the following article, Richard Kent, VP of Global Sales, VisionTrack explains how AI video telematics not only provides better safety in and around fleet vehicles, but will also monitor and analyze incidents as they occur to feed directly into improvements in fleet safety and a reduction of risk.

 Road safety is increasingly coming under the spotlight with various levels of government across Canada and the US. The latest statistics from the US National Highway Traffic Safety Administration (NHTSA) estimate 20,175 people died in motor vehicle traffic crashes in the first half of 2022, an increase of about 0.5% compared to 20,070 fatalities estimated by the NHTSA for the first half of 2021. The human cost is huge, but the additive cost to society of overall motor vehicle crashes is in the billions of dollars per year.

Safety policies are being put in place

At a city-by-city level, many communities have committed to strategies that are designed to eliminate traffic fatalities and severe injuries, while increasing safe, healthy, equitable mobility for all. There is a widespread consensus that many factors contribute to safe mobility – including roadway design, speeds, behaviours, technology, and policies – and clear objectives are being set to achieve a shared goal by engaging stakeholders that span local traffic planners and engineers, policymakers, and public health professionals. 

Technology aids road safety fleets need to get ‘on board’ with AI 

But policymaking alone will not deliver on the journey to reduce road deaths. It needs buy-in from fleet operators and the helping hand of some tech-led innovations. While we are not talking driverless vehicles yet, driver assisted technologies are certainly next on the road to full automation.

AI is already having an ever-greater influence on our everyday lives, so it is no surprise that it has a growing role to play within the fleet sector to help improve driver performance, support duty of care, and cut costs. Particularly in new developments, AI video telematics is expected to transform how vehicle operations approach road safety. In the broadest sense, AI is about using machines to perform tasks that would typically have required some form of human intervention and demonstrate behaviours associated with human intelligence. Powerful in-vehicle AI video telematics will make it easy to identify key areas of risk, reduce collisions and near misses, and ensure employees get home safely.

AI-enabled cameras go beyond the cab 

AI-powered vehicle cameras, using Advanced Driver Assist Systems (ADAS), Driver Status Monitoring (DSM) and Blind Spot Detection (BSD) technologies, are now enabling fleet operators to maintain safety levels for both their drivers and other road users. By automatically monitoring hazards on the road and high-risk behaviours, these devices make it possible to provide real-time feedback straight to the driver. 

Distractions such as cell phone use, eyes away from the road, smoking, eating and drinking can be detected alongside other fleet risks, such as fatigue, tailgating, and nearby vulnerable road users, so that drivers can be encouraged to change potentially dangerous habits. In fact, in one international deployment of AI-powered video telematics, installed across 16,000 public sector vehicles, there was a reduction in risky driver behaviour of over 80% within the first three months.

The latest intelligent detection cameras can even identify and track vulnerable people where driver visibility is poor, and risk of injury is high. These devices can establish the severity of risk dependent on the proximity to the vehicle of a worker, pedestrian, or cyclist, activating internal and external alarms when they enter virtual exclusion zones. This provides the driver with increased time to react and warns other road users of the potential risk. 

Humanized AI at work

Moving forward, advances in Vulnerable Road User (VRU) perception technology will enable AI-powered cameras to provide a nuanced understanding of human behaviour. Using machine learning techniques, it will be possible to train devices to accurately predict a person’s actions, thereby providing drivers with potential collision warnings that give them vital moments to avoid an incident. Backed by a dataset of hundreds of millions of human behaviours, the edge-based software analyzes age, direction, speed, and distraction to deliver a much higher degree of accuracy than traditional ADAS technology.

Real-time analysis and decision making when incidents occur 

Fleet managers can use the added insight provided by AI video telematics to better understand risk within their vehicle operations, and take steps to address issues before they result in a driving incident. However, no vehicle operation has the time and resources to manually review every triggered collision, near miss or driving event, when video uploads can exceed hundreds per day. Due to the size and weight of many vehicles – especially vans, trucks and specialist vehicles – dashcams require highly-sensitive g-force settings to detect a collision, which results in large levels of generated events data.

The now…

Computer vision algorithms can now be used to review huge amounts of data, which means fleet managers are only being presented with information that requires immediate intervention. For example, AI post analysis can help overcome the challenge of manually checking hours of downloaded footage by automatically validating in seconds whether a collision occurred and determining if any action is required. The technology will continue to evolve in the future to detect, monitor and analyze near misses and driver behaviour, which will support data-driven safety decision-making and problem-solving.

AI post analysis uses advanced object recognition software to identify different types of vehicles, cyclists, and pedestrians, making it possible to distinguish between collisions and false positives that can be generated by harsh driving, potholes, or speed humps. This added layer of analysis enables rapid intervention and the ability to quickly summon emergency assistance, resulting in enhanced duty of care and driver welfare, as well as reduced insurance claims costs.

The tech delivery

There are two types of technology – edge-based and cloud-based – that will see AI delivery become increasingly embedded in video telematics hardware and software. For edge-based solutions, the processing takes place close to the data source, such as a connected camera device, to provide real-time insight. Cloud-based solutions collect and process information in a centralized data centre for powerful post analysis.

Driving towards a safer future 

The new generation of AI video telematics will ensure fleet operators can access the right information at the right time, presented in a way that enables them to achieve significant change and encourage drivers to operate in the most responsible manner. By automating management processes, data analysis, and incident detection, they can take advantage of intelligent solutions to keep drivers, road users, and pedestrians safe, and reduce the number of collisions.

The U.S. DOT Fatality Analysis Reporting System (FARS) shows that large trucks account for nearly 13% of fatalities on the roads in the US, so there is an opportunity to embrace AI innovation and immediately save lives – and we must not ignore its ability to reduce cost to society as well. We all want a future where no one is killed or injured on our roads, and fleet technology such as AI has a significant role to play in safer transportation for all.