The Coming Robot Revolution, Part One: Driverless Cars
It is time to start writing on this blog again. All of my blogging energy has gone into stuandmags.net, and now that Mags and I have decided to settle down in Nevada City, California, it is time to do more writing, which is more what this blog is about.
Lately, I have been thinking more and more about the coming advances in robot and artificial intelligence technologies, and how much change these advances will bring about in our economy, transportation systems, and personal lives. My thinking has been stimulated and prompted by a piece by 60 Minutes on robots in the workplace, and a series of articles in Forbes about the Google Driverless Cars.
I think it is most clear what the impacts of the Google Driverless Cars will be, and that is the subject of this first post. Very quickly, this new technology will turn every part of our transportation system on its head.
First and foremost, in most respects driverless cars will be a huge boon to street safety for all users. Collisions between two driverless cars will be virtually non-existent, as the cars will be in constant communication with each other. The algorithms and sensors controlling the cars will have much quicker reaction times and much lower error rate than any human driver.
These algorithms will be required to follow all traffic laws. This will be a huge plus for all sorts of streets, but especially in cities. As a cyclist, I look forward to the days when cars will properly yield and merge, and not double park.
Even for the clueless, iPhone-wielding pedestrian will driverless cars be a great advance in safety. Even the best algorithms and sensors can’t bring a car to a dead stop in no time, but a driverless car could begin perfect evasive and slowing maneuvers the second a pedestrian steps into the road. This extra half second and use of perfect technique will avoid many collisions with all road users, no matter who is at fault, and when collisions do occur they will be much less likely to be fatal.
Perhaps the greatest advance in street safety will be the elimination of drunk driving. There will still be plenty of dolts who decide to drive themselves home when they have a driverless car option, but most drunks will opt to let the car drive them home. Drunk driving is a huge problem, as we all know, and I think it is a problem that will eventually be close to eliminated.
The entire car ownership model will be turned on its head. Many people will choose out of habit or pride to still own their own car, driverless or not, but more and more people will realize the great economic benefit of simply renting a driverless car on demand, paid hourly and/or by the mile. If you live in a city or suburb, you will be able to request a car to pick you up, and a car without anyone in it will drive itself to your front door within minutes. Most people will opt for this option over owning their own car, especially young people and city dwellers.
With near perfect utilization of a fleet of driverless cars, many fewer cars will need to be produced. This will be a giant shock wave to the auto industry. The vehicle fleet will eventually fall to 10% of what it is today, since cars will rarely be vacant, save for the times like 3am when most people don’t want to go anywhere anyway. This change alone will be disruptive to many economies, especially the United States, Japan, and Germany. A smart auto maker will realize this change is inevitable and get out ahead of the change. Half or more of current auto makers will fold within the next few decades as demand vanishes.
With far fewer cars sitting idle, giant parking lots will become a thing of the past, as will many areas of street parking. This will be a great thing for bicycle infrastructure, sidewalk widening, and using old parking lots for things like parks and public gardens. This will also cause business real estate prices to drop, as many huge parking lots will now be open to development for business use. Indeed, even streets themselves will be able to be significantly narrowed as driverless cars will be able to follow each other with just inches to spare. This narrowing and removal of car travel lanes will present even more opportunities for more sidewalks and bicycle infrastructure.
Again, driving will be much more efficient, and I mean an improvement by leaps and bounds. Gone will be the days of constant traffic jams due to irrational driving behavior, collisions, and inefficient use of road surfaces. Because driverless cars will be in constant communication with each other, they can follow each other within inches, braking instantly as needed, and gaining a huge boost in wind drag efficiency due to drafting. This huge increase in efficiency will make our current freeways look downright silly as most of the road surface will be completely empty while other parts will be crammed with cars safely going 100 miles per hour. Rather than our current ridiculous trend of widening freeways, we’ll actually contract them and save bundles on maintenance costs.
As you will be able to instantly get a personal vehicle sent to your location at any time, equipped with a computerized chauffeur, public transit will go through a huge change. Many public bus lines will be eliminated, and where they are kept the drivers will be replaced by computers and sensors. Some people will have their driverless car memberships subsidized by local government so the costs to them will be comparable to their current subsidized transit passes.
Other types of transit will show resiliency as they offer speed advantages over cars with rubber tires on asphalt and cement. Airplanes and eventually high-speed rail will be used when you need to get somewhere more quickly or more comfortably than a driverless car can get you there. However, routes like San Francisco to Los Angeles will greatly shrink as people can do the route in 3.5 hours using a driverless car at 100 miles per hour, without having to deal with the indignities and delays of airports. Air travel and high-speed rail will mostly be used for trips of 500 miles and more, and of course trips across the oceans. I don’t see good local trains and subways going away any time soon, as in many cases they are already such a fast and efficient way to travel.
Even many of our errands will be turned on their heads. You’ll be able to place an order on the web to the grocery store, and a car with no one in it will arrive at your house at the appointed time with your order. Your package deliveries will also arrive in a driverless vehicle at appointed times. A driverless car will arrive to take kids to school.
Taxi drivers will of course resist this entire endeavor, and of course I feel for them. But, honestly, resistance is futile. It is not a matter of if the taxi driver profession will be almost entirely wiped out, but simply a question of when.
Unfortunately, with the increases in efficiency, cost, speed, and convenience, many people will actually increase the amount of time they spend in cars. People won’t mind living two hours away from their job as much when they can work, read, eat, and drink from their moving car. Unfortunately this all equates to more time people will spend sitting on their asses, increasing incidences of diabetes, obesity, and heart disease.
Driverless cars will be the push that electric cars needed. When it is a computer rather than a human worrying about how, when, and where to charge, many companies will make the switch. Gasoline combustion engines will still have a significant role for longer trips, but for shorter trips electric cars will rule the day. This will be a significant advance in air quality and engine noise.
However, another form of noise will surely increase: that of cars moving at 100 miles per hour regardless of any engine noise. This background “wooshing” noise can travel for many miles, and I think it will increase as there are more cars moving at a faster speed at any given time.
Google already has working prototypes of the driverless car, and I think within three to five years we will start seeing them on the roads. In that time frame they may still be out of reach for most private owners, but they will start to be used for car sharing, where people can share the costs of these expensive cars, currently estimated to cost $300,000. The car sharing model will be the one that eventually comes to dominate, anyway. Costs for these cars will rapidly fall until they start to approach the price of current human-operated cars, and as the prices fall these things will really start to take off. I feel that most people have no idea what huge changes in transportation are on the horizon but soon enough everyone will know about it.
Despite this lengthy blog post detailing the many changes I expect to see, I am not necessarily a full advocate of driverless cars; I just think that these changes are inevitable. Overall, I think the changes will be a net positive over our current transportation system, mostly due to the huge advances in street safety for all users. However, we need to be aware of the significant drawbacks, some of which I have addressed above.
The biggest downside I see to driverless cars is that it will make it much more tempting to not engage in active transportation. Cycling, walking, and combining these modes with public transportation are currently attractive to many people because of the cost and convenience compared to private automobile ownership, especially in cities. As using driverless cars becomes inexpensive enough, many people will start to use the service and miss out on the significant health benefits that active transportation provides. In fact, I think cyclists and walkers will increasingly be seen as quaint and behind the times. The social and economic pressures to travel by driverless car will be great.
People who choose to not own a cell phone will probably be left in the dust, as they will be needed to get picked up anywhere you don’t have access to a computer.
As people are no longer in control of their own motion, they will lose touch with their sense of direction, even more than they already have with smart phones and GPS. People will lose the sense of connectedness of places as they will pay less attention to the geography between points A and B.
As active transportation decreases and public transportation shrinks, we will become more atomized as individuals than we already are. When you get picked up and dropped off door-to-door in an encapsulated vehicle, you interact less with people. The driverless deliveries will have the same effect as you no longer interact with people at the grocery store. However, you may be able to easily choose to carpool in a driverless car which would increase your interaction with people.
As it stands now, the easiest way to limit the adverse effects of car culture for yourself is to simply not own a car. It is a big, expensive decision to own a car and therefore you can easily stick with your original decision to not do so. However, when a car lifestyle is a simple web form and $50 membership fee away, it will be much more difficult to remain true to your decision to not own a car.
There will be impediments to the widespread adoption of the driverless car. These impediments will come in the form of cultural resistance, mostly from conservatives, resistance from the auto industry, and from lawmakers scared of change. These impediments are real, but at most they will slow but not stop the driverless car. Resistance is futile, for better or for worse.
This post has not been revised since publication.