• utkarshkej

The future of autonomous driving relies on improved line marking technology


Whenever you watch some futuristic movie from the 80’s and 90’s that’s set in what is now the present day, you can’t help but wonder why in 40 years we haven’t achieved some of things set out in those films.


Returning to the moon, stepping foot on Mars, and flying cars just to name some of the more obvious predictions that film writers have made!


Yet whilst Elon Musk’s SpaceX busies itself planning for settlements on Earth’s closest neighbour, we are certainly no closer to seeing a Vauxhall Corsa flying past the window on the Air1 or Air62 highway. And we haven’t really made much progress in the field of autonomous vehicles either.


There have been some tantalising glimpses and videos of concept vehicles that do hint to a future where a commute to work in a car could be no more taxing than sitting at home on the sofa. Pick your destination, sit back and relax, or even summon one from a fleet of shared cars to wherever you are when you need it. (Someone should trademark UberX if they haven’t already.)


That ultimate vision has given rise to some of the great tech we see in cars nowadays though. Lane assistance on motorways, and letting the car keep a set distance from the vehicle in front without having to touch the pedals. But that’s about it. We don’t appear to be any closer to a major car manufacturer releasing a pod-like room on wheels that can take you to wherever you want to go with little to no input from the ‘driver’.


Some key obstacles remain. Safety, first and foremost, is the key driver behind these projects, and autonomous vehicles just can’t be taught the nuances of driving that someone with 30 years on the road with a spotless no claims history can know. And when you get to the really intricate details, getting a car’s AI to understand when it’s safe to enter a roundabout when there are cars sat at each entry is more complicated the more you sit and think about it.


But there’s something even more fundamental that’s causing a big roadblock on autonomous vehicles, and that’s road markings.


Road markings are important to ensure safe and reliable navigation of autonomous vehicles through the roadway. Human drivers are good at spotting them, and of knowing where they should be if for whatever reason they’re not visible - like they’ve faded, have been covered by roadworks or are masked by inclement weather.


But autonomous vehicles need to see these line markings at all time to understand where they should be on the road and in relation to other road users.


As EuroRAP recently said; “But like the human eye, the technology cannot work effectively if it cannot see the road markings and traffic signs if they are worn out or hidden, or if they are confusing.”


Aforementioned Musk even noted that ‘we really need better lane markings in California’. CEO of Volvo, Lex Kerssemakers was less diplomatic when he stated: “It (autonomous car) can’t find the lane markings! You need to paint the bloody roads here!”


Herein then is the biggest problem facing the future of autonomous vehicles. It’s not the tech or the AI intelligence - that will eventually get to a place where it’s safe to use on the roads. It’s actually the line markings and road signage, because whilst advancements in road paint tech is ongoing, especially from an environmental standpoint, it’s even harder (and very expensive) to maintain all the line markings across all the world’s highways… all of the time.


Shop now: Line marking paint


There are 5 levels to autonomous driving. Level 0 is where most of us are, in complete control of the car. And level 2 is where some vehicles are some of the time, where the vehicle’s automated functions take care of some of the work. But level 3, where the driver is not required to monitor the environment at all, feels a long way off, partly due to the line marking issue.


And level 5 where the driver becomes the passenger? Well, we may see boots step foot on Mars before we see that really take off and become the norm.