Category Archives: robots

Google Self-Driving Car Project Vehicle

Redesigning places and networks for autonomous vehicles (Fast Roads – Slow Streets)

Google has been secretly working on a car. We knew that they were working on autonomous vehicles, but they have also been redesigning the car for an autonomous world and came up with a pod car. The design will be familiar with those who have been following Personal Rapid Transit, though an important difference is that it is in principle trackless (or rather the entire road network has been sufficiently mapped in detail so the whole world is track, rather than bespoke track).

The promotional video is below:

Google Self-Driving Car Project Vehicle
Google Self-Driving Car Project Vehicle

The newest vehicle is designed for slow speed (25 MPH) on campuses, and is especially light. The low mass is important as it saves energy but also causes less damage when it accidentally hits something or someone. Combining the low mass with the lower likelihood of a crash at low speed will magnify its safety advantage for non-occupants in this environment compared with faster heavier vehicles (which privilege the safety of the vehicle occupants).

While I had been assuming the first market for autonomous or semi-autonomous vehicles would be the relatively controlled environment of the freeway, the relatively controlled environment of low-speed places makes sense as well. These are two different types of vehicles (high speed freeway vs. low speed neighborhood), and though they may converge, there is no guarantee they will, and perhaps today’s converged multi-purpose vehicle will instead diverge.

There has long been discussion of Neighborhood Electric Vehicles, ranging from golf carts to something larger, which are in use in some communities, particularly southwestern US retirement complexes. In Sun City, Arizona, for instance, people use the golf cart not just for golfing, but for going to the clubhouse or local stores.

Golf Carts in Sun City Grande, Arizona
Golf Carts in Sun City Grande, Arizona

They can do this because local streets are set with low speed limits, and there are special paths where they are not.

How many places already fit this bill:

  • Campuses
  • Retirement communities
  • Neighborhoods in master planned communities
  • True parkways
  • Others?

Note that many of these places have gotten a bad rap from the current flavor of urban planning which decries non-gridded networks. However keep in mind that non-grids have the advantage of discourage through traffic. Perhaps roads are too wide or too fast in these places, but that is much easier to fix through traffic calming than a too connected network.

We will not only be able to deal with such ideal places. We will also need to do retrofits.
How many places could fit this bill:

  • Cities designed before the automobile, where the grid can be retrofitted to disallow high-speed traffic
  • Anywhere there is space to retrofit a slow network in parallel with the existing fast network
  • Others?

So will people buy such cars with limited speed? Many will as a second or third vehicle, as they already do with golf carts. The arguments are very similar to those about electric vehicles.

The opportunity arises with Cloud Commuting, when such cars, as they are autonomous, come to you. They will be dispatched when they are practical for the trip at hand, which may either be a short distance within a `slow space place’, or can travel along a `slow path’ between nearby places.

This slow path is of course faster than bike paths and sidewalks, but slower than Principal Arterials and freeways.

Retrofitting cities for transportation has a long history, cities and transportation co-evolve. We redesigned our cities, which had originally emerged with human and animal powered transportation, first for streetcars, and then for the automobile, and in some larger cities for subways. We have also redesigned our taller buildings for escalators and elevators.

We did not however redesign cities for Segways.

We have already differentiated speed on links, and setting speed limits is one of the key jobs of the traffic engineer in ensuring safety. This is not only on the link in question, but important for other links as well. Travelers shifted away from freeways and onto less safe rural roads when the speed limit was set to 55 MPH in the 1970s, and back when it was raised in the 1980s, improving overall safety, though not necessarily safety on the freeways themselves (See Lave and Elias 1994).

The issue continues to be debated:

If you need a sign to tell people to slow down, you design your street wrong (via Strong Towns)
If you need a sign to tell people to slow down, you design your street wrong (via Strong Towns)

Minnesota lawmakers demand higher speed limits on more miles

As part of an expansive budget bill signed into law last week, state lawmakers nudged transportation officials to boost the speed limit to 60 miles per hour on lane miles where it can “reasonably and safely” be done. By 2019, traffic engineers must examine every mile of road with a 55 mph limit and determine if it is prudent to go higher.

It’s an enormous undertaking. There are 6,771 miles on two-lane/two-way state highways now covered by a 55 mph limit. Officials figure they’ll get through about one-fifth per year, starting as soon as next month. They will analyze each stretch’s crash history, design, lane width, sight lines and ditch slope.

“The fact we’re studying the roads does not mean you can jump to the conclusion that all roads will be raised to 60 miles per hour,” said Peter Buchen, assistant state traffic engineer at the Minnesota Department of Transportation.

But the agency has been moving in that direction. In 2005, the department bumped the limit to 60 mph on 791 miles of two-lane highways and added another 750 miles last year. Buchen said those were prime candidates — straight, wide-open stretches with clear sight lines and low incidence of crashes. He said limits on hillier, curvier highways probably won’t budge.

So I will posit several Axioms about transportation

  • Axiom 1: Some roads should be fast – The aim of transportation is connecting people with destinations. They can connect with more destinations if they can do so in less time. Ceteris paribus, faster roads will take less time.
  • Axiom 2: Some roads should be slow – Some roads serve neighborhoods and have traffic that is not just motor vehicles. Ceteris paribus, slower roads are more likely to ensure safety, a high quality of life, and increased interaction within the neighborhood. Without loss of generality, let’s call these roads streets.
  • Axiom 3: Fast roads attract traffic from slow roads – In general, people prefer to spend less time traveling, and will spend less time on faster roads. These roads will attract more people. There will be net reductions in traffic on streets that are made slower and net increases in traffic on roads that are made faster.

We thus should redesign our road hierarchy with these axioms and the possibility of slow vehicles becoming mainstream, developing a slow network so that these neighborhood vehicles cannot not only travel within neighborhoods or on campuses, but between them.

Question: When will a robot car win a car race against human drivers?

One of the classic questions in Artificial Intelligence is when will a computer beat a master human in some game (Chess, Jeopardy). Well, the next test is not a simple board game, but something involving control of a physical device at high speeds. So the next test is when will a self-driving autonomous vehicle without a human in the drivers seat (at all or via remote control) beat the humans in a NASCAR, Indy, or Formula One car race?

My best guess is 2025, though I am fairly confident before 2029 (the bicentennial of the Rainhill Trials and a year before the bicentennial of the Tom Thumb on the B&O), but this is without a lot of basis. We don’t have enough data points of cars vs. humans in a race. The automakers are not going to be too keen on this, since it sends the wrong message perhaps, and their is a great deal of risk, so it will be after robo-cars are already somewhat widely deployed.

Obviously if you took the safeties off, and didn’t worry about crashing, a robo-car could drive faster than humans now on an empty track. But with other drivers, this would affect the computer, how would/should it respond?

Imagine: A World Where Nobody Owns Their Own Car – Eric Jaffe – The Atlantic Cities

Eric Jaffe at Atlantic Cities writes: Imagine: A World Where Nobody Owns Their Own Car :

“Skeptics have also charged that autonomous cars will disrupt any city-based travel models, since people freed from the need to drive will move even farther away from the core. That might be true for people who own their autonomous cars, says University of Minnesota transport scholar David Levinson, but a strong sharing system could promote the opposite movement. “If you’re paying for the car by the minute, then you’re not going to want to move farther out,” says Levinson. “You’re going to want to move closer in.”

Levinson says SAV service that offers convenient on-demand trips gives people a much greater incentive to rely on that system — and a much smaller incentive to take the unnecessary trips often made by private cars. It’s a recipe for the type of multi-modal lifestyle change only possible right now in places like Manhattan. Transit for essential daily trips, cabs (or other alternatives based on the type of trip) for the rest.”

The big market question is when this world will begin to emerge. Levinson subscribes to a timeline in which autonomous cars enter the luxury market in 2020, the technology trickles down into the affordable mid-level range over the next several years, and by 2030 every [edit: NEW] car on the road is driverless. (Other cars would be banned a decade later.) Since car- and ride-sharing operations tend to rely on smaller cars, that would peg SAV networks closer to 2030 — about 16 years from now.

“It’s not that far away anymore,” says Levinson. “But 16 years ago was 1998, and Google hadn’t been invented. So it’s a short time and it’s a long time.”

CNN 10 Ideas: Self-Driving Cars

I get quoted by Todd Leopold of CNN for one of their  10 Ideas: Self-Driving Cars

Illustration by Stewart Scott-Curran, CNN
Illustration by Stewart Scott-Curran, CNN

Yes, flying cars and portable jetpacks are still, for the most part, the province of science fiction.

But another popular sci-fi idea, self-driving transportation, is rapidly becoming reality.

It’s no secret that Google has been working steadily on an automated car for years. Indeed, it’s even testing a few models on the streets of California.

But, more quietly, both auto manufacturers and government entities are also jumping on the bandwagon.

Automakers are already equipping cars with sensors that know, for example, when you’re about to plow into the car in front of you and can brake accordingly. David Levinson, a civil engineering professor at the University of Minnesota who writes the Transportationist blog, believes partially automated cars could be hitting the market by the end of the decade.

“My guess is that there will be some stuff on the market by 2020 that will be automated in that you could probably do hands-off driving on freeways in specific situations,” he says.

However, he quickly cautions, that prediction comes with a number of caveats. Automakers are concerned about liability – after all, who’s at fault if an automated car gets in an accident? There’s also the matter of equipping roadways and signage with helpful technology, something car manufacturers don’t expect in the short term. (Let’s face it: It’s hard enough to get potholes fixed.)

Finally, self-driving cars will likely take a generation to reach critical mass, says Levinson. Just as electrics and hybrids are only now becoming part of the everyday fleet, expect the number of automated vehicles to grow slowly in their early years, while people get rid of their previous vehicles.

Nevertheless, they could provide a huge benefit to society. Delivery services such as FedEx and UPS could automate their vehicles. Urban dwellers, who already use services such as Uber and Zipcar, would have more options to get around. And self-driving cars would be safer, thanks to the kinds of sensors that are becoming widespread today, such as auto-braking and blind-spot recognition.

Of course, such advances take both political and financial will. Technology already exists to automate aspects of rail; systems in Europe and Asia (such as Japan’s bullet trains) are run by machine. Congress even passed a 2008 law pushing for the installation of positive train control (PTC), a technology that helps recognize dangerous conditions, but U.S. systems have been slow to implement it.

Such a system could prevent deadly railway crashes that appear to have been caused by human error, such as a July high-speed derailment in Spain that killed 79 people and the crash of a commuter train this month in New York City.

“We should be doing more automation,” says Levinson. “It’s a lot easier to automate rail systems that it is to automate cars and highways.”

But, gradually, we’re getting there, he says. Throw in other innovations – 3-D printing, which could eliminate the need to have certain items shipped; telecommuting, which is already creating “virtual offices”; and alternate energy sources, which may reduce dependency on fossil fuels – and 20th-century transportation styles may finally end up in the rear-view mirror.

Volvo “Drive Me” project brings 100 autonomous cars to the road

SlashGear reports: Volvo “Drive Me” project brings 100 autonomous cars to the road:

“Volvo today announced its new project ‘Drive Me’, which aims to bring 100 autonomous cars to Sweden’s roads, something that has been endorsed by the Swedish government. Under the project, Volvo and other authorities will ‘pinpoint the societal benefits’ of self-driving cars, as well as seeing them in use on average public roads in ordinary everyday driving conditions. The project will take place in the city of Gothenburg.”

H/T SZ

Automated Vehicles are Probably Legal in the United States

Bryant Walker Smith writes 99 pages saying Automated Vehicles are Probably Legal in the United States:

“This paper provides the most comprehensive discussion to date of whether so-called automated, autonomous, self-driving, or driverless vehicles can be lawfully sold and used on public roads in the United States. The short answer is that the computer direction of a motor vehicle’s steering, braking, and accelerating without real-time human input is probably legal. The long answer, which follows, provides a foundation for tailoring regulations and understanding liability issues related to these vehicles.
The paper’s largely descriptive analysis, which begins with the principle that everything is permitted unless prohibited, covers three key legal regimes: the 1949 Geneva Convention on Road Traffic, regulations enacted by the National Highway Traffic Safety Administration (NHTSA), and the vehicle codes of all fifty US states.
The Geneva Convention, to which the United States is a party, probably does not prohibit automated driving. The treaty promotes road safety by establishing uniform rules, one of which requires every vehicle or combination thereof to have a driver who is “at all times … able to control” it. However, this requirement is likely satisfied if a human is able to intervene in the automated vehicle’s operation.
NHTSA’s regulations, which include the Federal Motor Vehicle Safety Standards to which new vehicles must be certified, do not generally prohibit or uniquely burden automated vehicles, with the possible exception of one rule regarding emergency flashers.
State vehicle codes probably do not prohibit—but may complicate—automated driving. These codes assume the presence of licensed human drivers who are able to exercise human judgment, and particular rules may functionally require that presence. New York somewhat uniquely directs a driver to keep one hand on the wheel at all times. In addition, far more common rules mandating reasonable, prudent, practicable, and safe driving have uncertain application to automated vehicles and their users. Following distance requirements may also restrict the lawful operation of tightly spaced vehicle platoons. Many of these issues arise even in the three states that expressly regulate automated vehicles.
The primary purpose of this paper is to assess the current legal status of automated vehicles. However, the paper includes draft language for US states that wish to clarify this status. It also recommends five near-term measures that may help increase legal certainty without producing premature regulation. First, regulators and standards organizations should develop common vocabularies and definitions that are useful in the legal, technical, and public realms. Second, the United States should closely monitor efforts to amend or interpret the 1969 Vienna Convention, which contains language similar to the Geneva Convention but does not bind the United States. Third, NHTSA should indicate the likely scope and schedule of potential regulatory action. Fourth, US states should analyze how their vehicle codes would or should apply to automated vehicles, including those that have an identifiable human operator and those that do not. Finally, additional research on laws applicable to trucks, buses, taxis, low-speed vehicles, and other specialty vehicles may be useful. This is in addition to ongoing research into the other legal aspects of vehicle automation.”

(Via Marginal Revolution.)