The combination of AI and data from video surveillance systems is set to transform how we manage traffic and urban mobility over the coming years. From hyper tailored routing and vehicle-to-everything (V2X) communication, through to more effective infrastructure planning and measurement, these systems are laying the foundations for meaningful, important, and lifesaving progress.
By: Peter Abdelmassih Waller
Surveillance technology as an innovation platform
There is a fundamental shift happening with how we think about and use video surveillance technology. Advances in features like resolution, lowlight, and higher frame rates used to be about the viewer’s experience. Now, the quality, quantity, and breadth of data these systems capture means we can do so much more than simply surveil.
We can now achieve all manner of new and complex objectives, be they to react and respond to instances in real-time; anticipate, prevent and change outcomes; or even inform plans with new, varied data. It gives us more ways to gauge, measure, or even redefine success – particularly useful for the transport industry which has historically been confined by limited and narrow data.
Transport authorities and planners face an endless list of challenges: Increasing numbers of drivers, congestion, road wear and tear, emission targets, public health and safety, as well as worsening environmental patterns. It makes prioritizing and justifying investments particularly tricky, especially within shrinking departmental budgets.
Thanks to AI’s ability to process, analyze, and identify patterns within huge volumes of data, transport authorities will soon have far richer data to help inform these decisions, and a significant amount of it will come from surveillance technology. These systems are increasingly becoming a data platform on which exciting new use cases are being built which will shape tomorrow’s mobility. And the good news is that that there are already millions of cameras watching our roads and cities, generating data just waiting to be used.
Here are just a few of the near limitless ways surveillance-powered AI systems may transform our transport systems over the coming years:
Tailored route planning
There’s a lot more to route planning than you might think. While there have been improvements to account for traffic and alternative routes, these are applied with a blanket approach. Currently, there is little-to-no consideration of the vehicle type let alone broader, more conceptual objectives. Surveillance systems already provide real-time data of the driver’s vehicle (through visible features and license plate recognition technology) and could soon detail live road conditions. This makes them ideal resources to draw upon for more tailored route planning.
Weather, for example, can greatly affect vehicle performance. Wind speed and direction hugely impact fuel consumption. This is particularly pertinent to haulage and heavy goods vehicles which are already high fuel consumers and poorly streamlined. Electric vehicles, too. Lengthy charge times multiply any impact a reduction of range has on the journey. Likewise, icy or partially flooded roads may be traversable for 4x4s, but smaller vehicles and motorbikes need more suitable routes. Electric scooters are sensitive to poor road conditions, such as potholes, loose stones, or tram tracks. Even visibility, be it low light, fog, or other harsh weather could be factored in.
Real-time observations could also directly assist drivers and improve situational awareness. Surveillance systems might spot vehicles struggling with traction and suggest drivers lower their speed. Or pick up on subtleties such as road positioning or switching to fog lights.
Assisted travel and parking
Effective lane and speed management can be the difference between free-flowing traffic and miles long tailbacks. A system that monitors lane usage and advises which lane and speed to use, through in-vehicle infotainment and external signage, would be transformative. These same systems could also feed into traffic lights and signaling systems, prioritizing and rebalancing timings to the greatest effect. Likewise, incentivizing conformity to these instructions through some form of lower road tax or insurance premium would need reliable documentation, with cameras able to provide it. But these vehicles all have a destination, which means one thing: Parking.
Parking is a perennial issue for town and city councils. Up to 30% of drivers in congested urban environments are searching for a space. And this number doesn’t even account for those who have since given up looking and are travelling elsewhere instead. In short, any improvement here can have a huge knock-on impact on congestion and emissions.
Cameras can readily spot suitable spaces be they within a parking lot or on the street. Combined with traffic analysis and advances in V2X connectivity, drivers could be allocated and directed to spaces in the near future. The effectiveness of this only multiplies when you consider driverless vehicles communicating to one another who is parking where.
Similarly, these same systems could advise on engine type usage for hybrid vehicles. Knowing where and how many charging stations are available at a chosen destination, along with the levels of traffic and travel time needed, vehicles could be instructed to switch to and from battery powered modes.
Drivers could also be presented with alternative, more efficient options like parking further away and taking a shuttle bus in or renting an e-scooter or bicycle.
Improving road safety by identifying near misses
The primary goal of any transport authority is to make roads as safe as possible. Accidents and incidents are stringently reported and inform a lot of decision making. The issue with accident and incident tracking, however, is that near misses are not reported. Even if all signs point towards an imminent accident, it’s only after the fact that any steps are taken.
Surveillance data changes all of this. For high-risk areas, such as pedestrian crossings, roundabouts, and junctions, cameras and audio systems could readily track driver behavior on approach, spotting things like high speeds and rapid deceleration, skids, use of horns, and headlight flashes. This does two things. Firstly, it informs what kind of preventive measures should be taken, be that improved lighting, installing mirrors for greater visibility, or traffic calming bumps. And secondly, you have a wealth of new metrics to benchmark the effectiveness of any measures taken to improve road safety. Ultimately, this all results in safer roads with fewer accidents.
Redefining success for city and urban planning
With this ever-expanding pool of rich data, transport and planning authorities will soon make far more informed decisions. In fact, increasing numbers of towns and cities are investing in digital twins to model the efficacy of new changes and proposals before rolling them out. Surveillance systems are already proving an invaluable source of both historical and real-time data for this kind of planning. Pilot schemes and local initiatives can provide a feedback loop of learning, improvement, and verification ahead of major city-wide investments.
The results of all this testing and implementation could then be used to shape a variety of factors such as public safety or (re)investment policies, vehicle- or zone-specific rules, introduce new tax and revenue generating schemes, as well as new driver incentives.
Similarly new metrics for success can be introduced and readily monitored, such as reductions in near misses, decreased time to park/looking to park times, more consistent average speeds, all of which feed into wider public health and safety metrics beyond the individual department.
An instrumental contributor
We’re only just scratching the surface for what’s in store. But one this is clear: Just how instrumental surveillance data will be in enabling future traffic solutions. It all points towards transport and planning authorities having far more tools and levers to pull to make a better, safer, healthier, more sustainable and informed investments. Decisions will be more readily explained and justified to both the public and budget holders. And ultimately the overall experience for drivers, pedestrians, or otherwise will drastically improve.
