Currently reading: Autonomous cars could cut traffic in Britain by 40%, says report

New British research shows autonomous and connected cars can follow more closely and therefore make more efficient use of road space

A new study conducted by the UK’s Department for Transport shows that a full uptake of autonomous cars in Britain could cut road traffic by 40%.

The research concludes that connected and autonomous vehicles are able to follow each other more closely than human driven vehicles, so can, therefore, make more efficient use of space on the road.

UK road traffic at an all time high

Modelling urban and non-urban roads, the DfT computer programme was also able to gauge the effects autonomous or connected cars would have on journey times. It found that a road network of autonomous and connected cars would enjoy a 30.7% reduction in urban journey times during peak hours, while journeys on motorways and A-roads during peak hours would be 4.1% shorter.

The DfT research also simulated what traffic would be like if the UK road network had a mix of autonomous and human-driven vehicles. It found that traffic would reduce even if as little as 25% of cars were autonomous or connected.

Autonomous car insurance will be similar to normal policies, says report

Transport Minister Johns Hayes said, “This exciting and extensive study shows that driverless cars could vastly improve the flow of traffic in our towns and cities, offering huge benefits to motorists including reduced delays and more reliable journey times.

Driverless cars are just one example of cutting edge technology which could transform the way in which we travel in the future, particularly in providing new opportunities for those with reduced mobility. This study reinforces our belief that these technologies offer major benefits and this government will support their research.”

Along with the study, the DfT published a response to a consultation on insurance for driverless cars. The consultation aims to establish a model where insurers can cover the drivers of an autonomous car as well as the driverless vehicle itself.

British motorists don't trust driverless cars

Join the debate

Comments
11
Add a comment…
Bazzer 6 January 2017

Cut traffic by 40? Easy.

I have a simple idea that would cut traffic by 40%. Introduce a new rule that every driver has to undergo a driving test every five years. It would sweep millions of total incompetents off the roads, and those left would be skilled and able drivers. Half the drivers I know are lazy drivers, with bad habits, and poor observational and anticipational skill levels.
gavsmit 6 January 2017

Worrying

There are positives to autonomous cars I suppose, like taking the control of vehicles away from dangerous bad drivers, driving both legally and illegally.

But there are some considerable negatives too, in addition to killing the joy / freedom of driving and the love of cars in general. Also any research that comes from government departments fills me with cynical suspicion – not only for the competence of their analysis, but also their true political (or financial) objectives.

Aside from this, the main thing that worries me is the developed technology. I work in IT and am constantly appalled by substandard IT applications that are put live “within deadlines” and “on budget” but don’t actually work! Many applications aren’t tested properly, with the supplier expecting the poor customer to find all the bugs for them after go-live (this practice being a kind of involuntary beta testing).

So consider the extensive testing that an autonomous car would be required to go through to cover every single possibility of what could happen on a journey – that sounds impossible to me, as well as the developers still delivering the car on time and making a profit. But if a testing scenario was missed, and the car didn’t know what to do in a particular unexpected situation, what would happen? Something that could be easily avoided by a human driver could lead to a fatality if the autonomous car makes a sudden manoeuvre based on a bad decision caused by missing instructions on what it should do.

In addition, with so much reliance on technology now, it only takes a talented IT geek with an alliance to a terrorist group to hack the software and who knows what could happen. Imagine someone installing some malicious code into an autonomous car (maybe hidden within a regular ‘software update’ which will be required to keep things working or fixes to avoid the scenarios missed by insufficient testing mentioned above) which could suddenly trigger all cars to turn right sharply at the same time during rush hour on the M25?

Maybe advanced artificial intelligence can counter some of these worries, but I remain unconvinced. Autonomous cars are probably the most benefit to those who hate driving or are rubbish at it so shouldn’t be driving anyway. Those people have ruined motoring for all the rest of us.

spqr 6 January 2017

"You can nae change the Laws of Physics Captain"

As Commander Montgomery Scott once said. As pointed out above bunching cars up is not reducing traffic at all it is simply contracting the area it takes up. Superficially this sounds as though road space will be freed up. But who is then going to use the additional road space? For decades the DfT mantra has been "building more roads leads to more traffic" so they will not build roads. Now the DfT says autonomous cars will create more road space surely applying their old mantra this means more traffic as the effect is the same as building extra road space? There are also implications for safety if cars are bunched up. For example some drivers may not be as fastidious as others when having brake pads and discs serviced. You know the sort - there is a further 500 miles in left in the disc/pad so leave it until then to get it replaced. So assuming there is a sudden need for the "convoy" of autonomous cars to brake suddenly some will manage the task some will not. That means, as physics (and Commander Scott) tells us, some cars will collide with the others as the stopping distances are too short. One could also look at this issue from the point of view of fractal mathematics (small variations in the software/hardware having huge impacts further down the line) but therein lies chaos and ruin for the DfT and anyone else championing autonomous cars.