The Case for the Driverless Car

David Murphy

Mercedes-Benz F015 - Luxury in Motion

Driverless cars are an emotive issue. Personally, I’m a big fan, partly because they seem so much safer than leaving humans in control. But mainly because, faced with the prospect of a 5- or 6-hour drive up the M40 and M6 to see my family up North on a Friday evening, as I am a few times a year, the idea of sitting back and doing something more interesting or relaxing than staring at white lines and bumpers for 300 miles just seems to me like a no-brainer.

But not everyone you talk to agrees. In fact, the main reasons people tell me they don’t like the idea of driverless cars are exactly the same as the reasons I do like them. Firstly, they say, they can’t be safer than having a human in control, can they? And secondly, they say, I actually enjoy driving. Why would I want to deny myself the pleasure?

Transition period
For what it’s worth, I think that for two or three decades from 2020 or so onwards, we’ll see a transition period where people still have a car they can call their own, but can summon up a driverless car for those long journeys where they want to do something better with their time than concentrate on the road ahead.

And just think how a driverless cab service could work. You summon the cab to get you from A to B and the app tells you the journey will cost you £10. Halfway down the route, two more people looking to get to the same place get in and the fare you have to pay drops accordingly. A great idea and perfectly achievable.

Mercedes-Benz F015 - Luxury in Motion

Interior Design - the inside of Mercedes' F015 concept driverless car

And don’t be fooled by the fact the first incarnations of the driverless car look like something out of the Jetsons. Car makers like Mercedes Benz are already envisioning the day when if no one needs to drive the car, there’s no need for it to be laid out as it is now, with four seats all facing forwards. Mercedes’ F015 concept car offers a glimpse into what this future might look like, with two pairs of seats facing each other, enabling the occupants to hold a business meeting or play a board game while the car takes care of driving itself. And if you accept this, then why not have a driverless car the size of a transit van that doubles as a home cinema or a gym to while away those long hours on the motorway. Or for a here-and-now take on it, take a look at the Tesla, of which more later.

Compelling case
At the recent Internet of Things Forum in Cambridge, Professor John Miles from the University of Cambridge’s Department of Engineering made a pretty compelling case for the driverless car. Anyone who has ever been stuck in a traffic jam, he noted, will have recalled the case that is often made for getting some of the traffic, notably freight traffic, off the roads and onto the railways. After all, he explained, over 90 per cent of passenger kilometres in the UK are delivered on the road, with rail accounting for just 8 per cent. For freight, 68 per cent of freight kilometres are on the road, with just 9 per cent are on the railways, and shipping making up the balance.

The problem with the argument of moving more stuff off the roads and onto the railways, however, is the infrastructure. The UK has 224,000 miles of road network, of which 29,000 miles are motorways and main roads. When it comes to rail, there are a mere 10,000 miles of railway tracks in the UK, so even if the railways could relieve some of the pressure on the road network, the passengers and freight would still have to revert to the roads for the last mile, or few miles.

So if the railways are not the answer to congested roads, what is? According to Professor Miles, the driverless car. On the face of it, he said, there does not seem to be much headroom on the road network, given the number of traffic jams that drivers encounter on a daily basis. But 30 per cent of the congestion on our roads, he said, is caused by minor incidents, and 80 per cent of these are caused by driver inattention. “It’s not difficult for a machine to beat an inattentive human,” he concluded.

Indeed, last July, in the aftermath of an incident in which a Google driverless car was rear-ended by a car under human control, Google confirmed that the shunt was only the 14th accident in six years and just short of 2m miles of testing. 11 of the 14 involved the driverless car being rear-ended, while the only one in which the driverless car was at fault happened when it was under human control.

Professor Miles then went on to explain how driverless cars could reduce congestion in other ways. If the car is in control, and you can trust the car not to crash into the car in front of it, then it can drive closer to it, from both a front-to-back and side-to-side perspective. In effect then, a three-lane motorway could become a four-lane motorway, and by driving closer to the car in front, he argued, a 25 per cent increase in headroom on main roads, and a 50 per cent increase on motorways, is not unreasonable.

I put it to Professor Miles that his vision sounds great and might well be achievable 30 or 40 years from now when all cars are driverless (perhaps). But what about the interim period, when normal people driving normal cars see the driverless cars getting very close to each other and think they can do the same, resulting in more of those minor – and major – incidents which create so much congestion? A fair point, he conceded, adding that during this transitional period of several decades, safety, rather increasing headroom, must be the key consideration.

In the meantime, there are some interesting tests going on around driverless vehicles of all shapes and sizes such as the L-SATS (Low-speed Autonomous Transport Systems) designed to get people the last mile from the train station to their home, currently being tested in Milton Keynes. Or the Bullet, a driverless, 120mph, 46-seat vehicle designed to get people to the edge of cities on its own dedicated expressway. It can be coupled with other Bullets to provide increased capacity at peak times and less capacity at quieter times.

“We will need lots of different systems to complete one journey, but it will rely on reliable information coming to your phone,” said Professor Miles. This information, he said, would include a variety of factors. Someone trying to choose where to go shopping could call up information on everything from the current traffic and public transport status, to how busy the shopping centre is, to the availability of car parking spaces. And of course, as he notes, the phone is at the heart of all this, especially if an alert you receive en route enables you to change your plans because of circumstances beyond your control.

How all this pans out remains to be seen. But there is money to be made in making the driverless car and the ecosystem that goes with it a reality, which history shows means it will tend to happen. And the genie is already out of the bottle. Ask my friend Harold, whose Tesla is more than capable of driving itself on any road with white lines in the middle, parking itself in tight spaces, and even getting itself out of the garage and onto his drive in the morning ready for the day ahead, while Harold enjoys a morning cup of coffee and catches up with the news on his iPad. Now does the driverless car make sense?