More on Google’s self-driving car program

Stanford’s Sebastian Thrun is my hero for steering the Stanford DARPA Urban Challenge project to success. Thrun is one of the hired-guns on Google’s high intensity project (think a bit on the technologies that might enable robo-cars: the synergy of AI + mobile + localization + GIS database). I am very curious how much of the Google robo-car algorithms depend on real-time data coming down from their cloud. Nobody can match Google’s cloud for auto-driving:

(…) So we have developed technology for cars that can drive themselves. Our automated cars, manned by trained operators, just drove from our Mountain View campus to our Santa Monica office and on to Hollywood Boulevard. They’ve driven down Lombard Street, crossed the Golden Gate bridge, navigated the Pacific Coast Highway, and even made it all the way around Lake Tahoe. All in all, our self-driving cars have logged over 140,000 miles. We think this is a first in robotics research.

Our automated cars use video cameras, radar sensors and a laser range finder to “see” other traffic, as well as detailed maps (which we collect using manually driven vehicles) to navigate the road ahead. This is all made possible by Google’s data centers, which can process the enormous amounts of information gathered by our cars when mapping their terrain.


Read more »

Google self-driving cars

Charles just sent us the link to this NYT Science article on Google’s self-driving car program. We’ve been following the Stanford-Volkswagen program — I did not know that Google was participating. Is Sebastian Thrun now a Google employee? Sebastian lead Stanford’s “Stanley” DARPA Grand Challenge project.

(…) During a half-hour drive beginning on Google’s campus 35 miles south of San Francisco last Wednesday, a Prius equipped with a variety of sensors and following a route programmed into the GPS navigation system nimbly accelerated in the entrance lane and merged into fast-moving traffic on Highway 101, the freeway through Silicon Valley.

It drove at the speed limit, which it knew because the limit for every road is included in its database, and left the freeway several exits later. The device atop the car produced a detailed map of the environment.

The car then drove in city traffic through Mountain View, stopping for lights and stop signs, as well as making announcements like “approaching a crosswalk” (to warn the human at the wheel) or “turn ahead” in a pleasant female voice. This same pleasant voice would, engineers said, alert the driver if a master control system detected anything amiss with the various sensors.

MIT Smart Cities: City Car

When I think of car sharing in the U.S., I think of Flexcar and Zipcar [they have merged — new name is Zipcar]. Together I think they had around 5,000 vehicles at merger time in 2007. But so far carsharing has no measurable impact on urban traffic or CO2 load. Some of the consumer resistance may be price. That’s where the MIT City Car looks promising — this is exactly what I want for the urban short-trips that involved carrying stuff back to the transport station. I could imagine the Buenos Aires CBD traffic density being cut in half or more by a hundred thousand of these way-cool cars, with a “luggage cart” stand every couple of blocks. And a little car that can move in any direction on its four independent wheel-robots would be very appealing — though possibly deadly amidst speeding BA taxis…

UPDATE 080212: Some clarifications are required, prompted by comments to this post from carsharing pioneer Dave Brook — whose
blog is a recommended source on the industry. I’ll just briefly enumerate these points:

1. Carsharing cannot make a major impact on urban traffic/CO2 load unless the concept wins large scale adoption by consumers who elect to substitute public transport + carsharing for their current single-passenger-per-vehicle preferences.

2. Not being privy to any objective studies of consumer preferences, I’m speculating that at least two factors will gate acceptance: convenience and price. On convenience the City Car concept
might contribute if deployed with sufficient spatial frequency to e.g. make Ms. shopper happy running her errands within a convenient carsharing radius around a transport station. My speculation is that density is probably not coarser than a 2 block grid.

3. Price — The City Car could make high frequency deployment more economic by introducing several operating cost efficiencies: lower real estate footprint; lower maintenance; lower cost per mile; lower labor cost assuming the automated kiosks are successful.

4. Pricing model — Both carsharing and the City Car concept share a pay-for-use pricing model. Motivating consumers to try out a substitution as radical as “drive myself” replaced by “public transport” surely will be facilitated by pay-for-use, avoiding the requirement of a new capital investment by the consumer. So I don’t see the slow acceptance of NEV’s, noted by Dave, as being an indicator of the future of carsharing options.

5. Price & Volume — another laudable MIT-based project, OLPC, has run into real-world complications with achieving low costs projected for high volume production. My understanding is the OLPC problem isn’t so much with the accuracy of forecasting manufacturing cost at XX units/year rates. Rather it is when early actualized demand turns out to be only 10% XX units/year, which then feeds into buyer resistance to higher costs –>> in to realized volumes. A non-virtuous circle. The City Car design looks like it could lead to an efficient high-volume design — if the capital partners are prepared to take the up front risk.

6. Not to prejudice your conclusions, but I suspect the success of the City Car concept will be gated by the vision and risk appetite of any group willing to undertake such a capital-intensive deployment. I don’t see how it can be crawl-walk-run — the strategy that has been successful for the carsharing pioneers. Can it launch with only one suitable city? Yes, if investors are willing to accept the risk of the required volume commitment turning to custard.

7. Lastly, keep in mind that the existing competition for the City Car concept is the taxicab industry — think about what has to be offered that will entice you to switch.

8. Please keep an eye on the Wikipedia page on the City Car.

The City Car is a stackable electric two-passenger city vehicle. The one-way sharable user model is designed to be used in dense urban areas. Vehicle Stacks will be placed throughout the city to create an urban transportation network that takes advantage of existing infrastructure such as subway and bus lines. By placing stacks in urban spaces and key points of convergence, the vehicle allows the citizens the flexibility to combine mass transit effectively with individualized mobility. The stack receives incoming vehicles and electrically charges them. Similar to luggage carts at the airport, users simply take the first fully charged vehicle at the front of the stack.

The City car is NOT a replacement for personal vehicles, taxis, buses, or trucks; it is a NEW vehicle type that promotes a socially responsible and more effective means of urban mobility.

The City car utilizes fully integrated in-wheel electric motors and suspension systems called, “Wheel Robots.” The wheel robots eliminate the need traditional drive train configurations like engine blocks, gear boxes, and differentials because they are self-contained, digitally controlled, and reconfigurable. Additionally, the wheel robot provides all wheel power and steering capable of 360 degrees of movement, thus allowing for Omni-directional movement. The vehicle can maneuver in tight urban spaces and park by sideways translation. This technology is patented-pending and under design development at the MIT Media Lab.

Here’s a very short news-blurb explaining the “last mile” challenge, which the City Car concept might address.

“The problem with mass transit is it kind of takes you to where you want to go and at the approximate time you want to get there, but not exactly. Sometimes you have to walk up to a mile from the last train or subway stop,” said Franco Vairani, a Ph.D. candidate at MIT’s school of architecture. The City Car is his thesis, though it’s now a group effort involving many others at the school.

I don’t respect The Guardian, but they seem to have the best popular coverage of the project:

It is not every day that a concept car re-writes the rules of more than 100 years of motoring. In development for four years by a team of architects and engineers led by William Mitchell, former head of the school of architecture at the Massachusetts Institute of Technology (MIT), as part of his Smart Cities research group, a new MIT car is borne of a complete rethink of people’s relationship with their cars in the ever-expanding cities of the future.

Prof Mitchell expects we will share cars that will be easier to drive in congested cities, will be pollution-free and can be customised at will.

The city car concept, with styling input by architect Frank Gehry, will be completed and delivered by MIT to General Motors early next year.

“Primarily we’re interested in urban living,” says Ryan Chin, an architect and engineer at MIT’s media lab and a member of Prof Mitchell’s research group. “Everything scales down from what we think the city of the future is.”

The Smart Cities group focused on how cars could be better adapted to get round familiar problems of city life, namely congestion, pollution and parking. Motor companies are well aware of the issue. But the group felt the companies had missed the point, even with city cars such as the Smart, the iconic two-passenger cars introduced by Swatch and Mercedes in 1998.

“We have to think of city cars as not just small-footprint vehicles that can squeeze into tight spaces but ones that can work in unison and also be almost like a parasite that leeches on to mass-transit systems,” says Mr Chin. While Smart changed the way people think about parking and size, the MIT engineers felt that, as it had not been widely adopted and congestion and pollution problems had got no better, its success had been limited.

So the MIT team started from scratch to come up with their own concept: a stackable, shareable, electric, two-passenger car. “Imagine a shopping cart – a vehicle that can stack – you can take the first vehicle out of a stack and off you go,” says Mr Chin. “These stacks would be placed throughout the city. A good place would be outside a subway station or a bus line or an airport, places where there’s a convergence of transportation lines and people.”

[more Guardian]

Lastly, a very cool MIT poster on the City Car [PDF].