I want this technology to succeed, like … yesterday.
I'm biased. Earlier this year my mom had a stroke. It damaged the visual cortex of her brain, and her vision was impaired to the point that she'll probably never drive again. This reduced her from a fully-functional, independent human being with a career and a buzzing social life into someone who is homebound, disabled, and powerless.
When discussing self-driving cars, people tend to ask a lot of superficial questions: how much will these cars cost? Is this supposed to replace my car at home? Is this supposed to replace taxis or Uber? What if I need to use a drive-thru?
They ignore the smarter questions. They ignore the fact that 45% of disabled people in the US still work. They ignore the fact that 95% of a car's lifetime is spent parked. They ignore how this technology could transform the lives of the elderly, or eradicate the need for parking lots or garages or gas stations. They dismiss the entire concept because they don't think a computer could ever be as good at merging on the freeway as they are.
They ignore the great, big, beautiful picture staring them right in the face: that this technology could make our lives so much better.
Feelgood Britain is becoming gridlocked Britain. Fears of a direct link between a return to economic growth and ever-worsening traffic congestion are borne out by new figures from the Department for Transport which show traffic slowing to a crawl as sales of cars and lorries start to pick up.
During the recession, average speeds across the country’s road network rose despite steep cuts in infrastructure spending. They are now back below 25mph for the first time since the 2008 financial crash. Average speeds in central London often fall below 10mph. They are not much better in Birmingham, Bristol, Manchester or Newcastle, and without radical new thinking from ministers and urban planners they will get steadily worse. Despite the opportunities created by the internet for working from home, overall road usage is expected to grow by three quarters by 2040.
Traffic jams waste fuel, time and money on a prodigious scale. They hurt productivity, drive up air pollution and damage Britain’s image and competitiveness as global business powerhouse.
The Conservatives and Labour have promised sharply increased road-building during the next parliament, but the obvious solutions will not be enough. The spectacle of the two main parties outbidding each other on plans for new bypasses and crawler lanes already looks retrograde. The most important infrastructure contest of the next five years will not be national, but international. It is a contest in which Britain is currently an also-ran. It will not be about building more roads but moving cars along them faster and more efficiently — by letting computers do the driving.
When Google first publicised its work on driverless cars in 2005 they were seen as a faintly comical diversion for a company with more money than it knew what to do with.
Since then driverless technology, consisting of computer-linked cameras, radar and lasers, has developed to the point that it can safely steer a car past a wobbling cyclist, as a test fleet showed in California this week. Driverless cars can slip smoothly through complex junctions, always obeying traffic lights, and boost motorway capacity five-fold by allowing cars to bunch much closer together at high speeds than would be safe with human drivers.
It is now clear that driverless cars will be safer, not less safe, than driven ones. Studies suggest they could save tens of thousands of lives a year in the United States alone. At $150,000 per car the technology is still expensive, but costs are expected to fall thanks to economies of scale as more US states join California, Arizona and Texas in legalising testing on public roads.
Almost none of this testing is happening in Britain. David Cameron has said that he wants this country to be a world leader in driverless technology, but talk is cheap. Britain’s leading contender in the field is the University of Oxford’s Mobile Robotics Group, which this week announced plans to build three two-person prototypes capable of travelling at the less-than-mind-bending speed of 7mph. Meanwhile, BMW, Toyota and Nissan are in a serious race with Google to bring full-size driverless cars to the market within three years.
The next big disruptive technology will transform commuting and save billions of working hours a year. Those who profit from it most will be those who build it as well as use it. If Britain is to join this group, it must change gear.
Why do you actually need a car, anyway?
Computer Weekly November 14, 2014
At a recent BMC event, CEO and Chairman Bob Beauchamp stood on stage and gave a view on how the rise of the autonomous car could result in major changes in many different areas.
The argument went something along these lines - as individuals start to use autonomous cars, they see less value in the vehicle itself. The "driving experience" disappears, and the vehicle is seen far more as a tool than a desirable object. By using autonomous vehicles, congestion can be avoided, both through the vehicles adapting to driving conditions, accidents being avoided, areas where non-autonomous vehicles are causing problems being by-passed and so on. The experience becomes an analogue to SDN - the car's function can be seen as the data plane (it gets from point A to point B) is decided by a set of commands (control plane deciding what should happen) through commands issues by the management plane (what is the best way to get from point A to point B?).
It is then seen that the tool is not being used that much - for long periods of time, it is in the garage, drive or roadway doing nothing. It needs to be insured; needs to be maintained - it becomes an issue, rather than a "must have".
Far better to just rent a vehicle as and when you need it - a "car as a service" approach means that you don't need to maintain the vehicle. Insurance is a moot point - you aren't driving the vehicle anyway; it is the multiple computer "brains" that are doing so, working a full 360 degrees at computer speed, never getting tired; never failing to notice and extrapolate events going on around them. Insurance is cheaper and only has to cover damage caused by e.g. vandalism and fire: theft is out, as the vehicle is autonomous anyway and can be tied in to a central controller.
Insurance companies struggle; car manufacturers have to move away from marketing based on seeing fast cars driving on deserted roads to selling to large centralised fleet managers who are only interested in overall lifetime cost of ownership. Houses can change - no need for a garage or a drive and cities can change with less need of parking spaces. More living space can be put in the same area - or more properties on the same plot of land. Autonomous driving means less time spent commuting; less frustration; less fuel being used up in stop-start traffic.
When Bob first said this, my immediate response was "it will never happen". I like my car; I like the sense of personal ownership and the driving experience that I get - on an open road.
However, I then took more of an outside view of it. Already, I have friends in large cities such as London who do not own a car. They use public transport for a lot of their day-to-day needs, and where they need a vehicle, they hire one for a short period of time. Whereas this may have been on a daily basis via Hertz or Avis in the past, newer companies such as City Car Club allow you rent a vehicle by the hour and pick it up from a designated parking bay close to you and drop it off in the same way wherever you want. The rise of Uber as a callable taxicab company is also showing how more people want the ease of using a car, but not in owning the vehicle themselves. These friends have no requirement for a flashy car badge or for the capability to get in "their" car and drive it at any time - in fact, the majority do not like driving at all, and would jump at the chance of using an autonomous vehicle, so removing this last issue for them.
As tech companies like Google improve their autonomous vehicles on a rapid basis, manufacturers such as Mercedes Benz, Ford and GM are having to respond. Already, over fifty 500 tonne Caterpillar and Komatsu trucks are being used in Australia to move mining material, running truly autonomously in convoys across private roads in the outback, allowing 24x7 operations with lower safety issues.
Just as the car manufacturers are coming out of a very bad period, they now stand a chance of being hit by new players in the market. Elon Musk, of Tesla electric car fame, is a strong proponent of autonomous vehicles. Amazon would like to take on Google, and it is likely that other high-tech companies will look to the Far East for help in building simple vehicles that can be used in urban situations via a central subscription model.
Sure, such a move to a predominantly autonomous vehicle model will take some time. There will be dinosaurs such as myself who will fight to maintain ownership of a car that has to be manually driven. There will be the need to show that the vehicle is truly autonomous; that it does not require continuous connectivity to a network to maintain a safe environment. More companies such as City Car Club will need to be brought about, and suitable long-term business and technology models put in place to manage large car fleets and get them to customers rapidly and effectively without a need for massive acreage of space to store cars not being used. Superfast recharging systems need to be more commonplace; these vehicles need to be able to recharge in minutes rather than hours, or to use replaceable battery packs.
Certainly, moving to the use of autonomous electronic vehicles where overall utilisation rates can be pushed above 60% would result in far less congestion in city centres and so in less pollution, less impact on citizens' health and less time wasted in the morning and evening rush hours. Indeed, Helsinki has set itself a target of zero private car ownership by 2025.
At the current rate of innovation and improvement in autonomous vehicles, it is becoming more of a "when" than an "if" as to when we will see a major change in car ownership. The impact on existing companies involved in the car industry cannot be underestimated. The need for improved technology and for technology vendors to work together to ensure that an autonomous future can and will happen is showing signs of being met.
Teen Drivers Who Shouldn't Be On The Road
There's a new urgency to bring Google's self-driving cars to consumers, and for one of the program's top executives, it's getting personal.
The head of the company's driverless car program, Chris Urmson, told attendees at the TED conference here that his 11-year-old son could get his driver's permit within the next four and a half years.
"My team and I are committed to making sure that doesn’t happen," he said.
Google is aiming to get its driverless cars on the streets by 2020. Whether it's able to do so is unclear. It's been working on plans for an autonomous vehicle for years, but questions remain about whether government regulators will allow the vehicles or if consumers will pay for them.
The driverless car space is also becoming increasingly competitive. Major automakers like Mercedes, Audi and BMW are rushing to develop models. Apple is rumored to be working on a car of its own.
Urmson's talk, part of a larger discussion at TED about advances in machine learning, didn't address competitors, but instead focused on how Google's cars operate.
Visuals of test drives wooed the crowd. The cars can maneuver around traffic cones, pedestrians, and even obstacles that would otherwise be hidden from the driver, like an out-of-view cyclist about to run a red light.
In one particularly colorful example, Urmson showed data from a self-driving car in Mountain View, Calif. as it tried to figure out what to do as a woman in a motorized wheelchair chased a duck in circles in the center of the road. (The car stopped and waited for the woman to move.)
"There's nowhere in the (Department of Motor Vehicle's) handbook that tells you how to deal with that," Umson said to laughs.
Tesla CEO Elon Musk gave his own endorsement of self-driving cars at a separate event Tuesday in California. He called the technology a "solved problem" and predicted that such cars would become commonplace in the next 20 years.
"In the distant future people may outlaw driver cars," Musk said, according to a Wall Street Journal report of the event. "You can’t have a person operating a two-ton death machine!”
Musk then said Tesla would become a leader in sales of autonomous cars.
Other automakers, like Ford, are focusing on technology that can assist human drivers by, say, tapping the brakes in traffic or helping with parallel parking. But Google's Urmson made the case that automakers must go one step further in creating cars that require no involvement from a driver.
His reasoning: About 1.2 million people are killed on the world's roads each year, with 33,000 car-related deaths in the United States alone. (That's the equivalent of a Boeing 737 falling out of the sky "every working day," Urmson said.)
"We think the right path is self-driving because the urgency is so large," he said.
Old Drivers Who Shouldn't Be On The Road
Tens of thousands of drivers with dementia are still on the road despite posing a risk to the safety of others because of flaws in the licensing system, a senior doctor has claimed.
Britain’s increasingly crowded road network and ageing population mean that it is only a matter of time before a dementia patient causes a catastrophic accident, according to a leading figure at the British Medical Association.
Peter Holden said that many drivers whose faculties were fading covered their condition up by “rote-driving” techniques such as avoiding right-hand turns and motorways, but were still a danger to themselves and others.
When drivers reach the age of 70 their licences automatically expire and they must renew them every three years. They face a heavy fine if they fail to declare a diagnosis of dementia, which does not necessarily rule them out from continuing to drive unless their doctor intervenes.
Dr Holden, a GP who sits on the BMA’s council and tabled the proposals, said that many dementia patients could complete all the individual manoeuvres in the driving test and yet posed a serious threat.
He said that many drivers had adopted coping mechanisms such as taking the same few routes or avoiding night driving, but they were unfit to deal with sudden challenges.
“My grave concern is that one day we are going to be faced with something like that and somebody will plough a car into a line of children,” he said.
“Boxes are being ticked but the dots are not being joined up. Driving is a complex activity requiring many faculties, and with dementia you cannot integrate them.” While you may tick the boxes for having adequate vision, adequate hearing, adequate musculoskeletal abilities, you may not have the brain power to put them together. He added: “You can find yourself in a position where you are no longer fit to drive even if you are technically deemed fit to drive. Everybody regards a driving licence as a right. It’s not. It’s a privilege.”
Today the BMA conference in Liverpool will vote on whether to ask its science board to investigate “the increasing problem of the potential impairment of judgment of some elderly drivers, with particular reference to those in the early stages of dementia”.
Dr Holden said he expected the motion to pass and that the review would recommend a change in the law by the end of the year. He called for the medical forms of the Driver and Vehicle Licensing Agency (DVLA) to be updated with the symptoms of dementia.
“The system hasn’t kept up with society, with the huge increase in people who are at an age where dementia is not uncommon,” he said.
“I had one patient [who] drove the car but his wife changed the gears because he couldn’t manage the gear lever.”
Although dementia patients are obliged to declare their diagnosis to the DVLA, Dr Holden said many did not do so because “turkeys don’t vote for Christmas”.
More than 800,000 people in Britain have dementia diagnoses, according to the Alzheimer’s Society. Dr Holden said that in a typical year each of Britain’s 10,000 or so GP surgeries might see five or six patients whose dementia rendered them unfit to drive. He also claimed that transport ministers had commissioned a report on drivers with dementia a decade ago but had buried their findings because they were “frit”.(Brit slang for frightened, apparently)