Alphabet Creating Stand-Alone Self-Driving Car Business | Hacker News Reader

Alphabet Creating Stand-Alone Self-Driving Car Business

wsj.com

71 pointsjbredeche34 comments

Out of curiosity, has anyone seen Google's self-driving cars testing down El Camino, University or Castro in PA and MV?

Each street presents uniquely complicated obstacles. In the case of El Camino, the police recently shut down the lights at an intersection due to Stanford game traffic. I'm curious if the self-driving cars know how to recognize hand signals from a legit police officer.

Over on University and Castro, there is a lot of foot traffic and sometimes people briefly double parking to drop off/pick up people. Some human drivers in these cases will (illegally?) cross the double yellow lines into the oncoming lane to pass the double-parked vehicles. Again, I wonder how the self-driving cars handle these scenarios.

In my experience they don't really stand out other than being a bit slow (for which it actually got ticketed).

They seem to just be very conservative.

The car did not actually get ticketed for driving too slow, it was only pulled over.

http://www.mercurynews.com/2015/11/12/mountain-view-google-s...

I wonder how a fully-self driving car would handle being pulled over. Could be like Robocop but in reverse (i.e., human police escalating due to non-responsive robot).

I'm also wondering about how portable the machine learning for the self-driving cars are.

Driving around, say, Europe, differs in subtle but important ways (no right on red, different on-pavement and side signaling, etc.) that it looks to me like they'd have to re-learn many of the aspects.

For example, Musk recently commented about how the Tesla car recognized what a valid parking spot was. That's totally dependent on location (country).

> Some human drivers in these cases will (illegally?) cross the double yellow lines into the oncoming lane to pass the double-parked vehicles

This is legal in my country, I mean, crossing the continuous white line in case there's an obstacle blocking your advancement, in this case the obstacle being the parked cars you mention. Granted, there still are obtuse (I'd say stupid) cops in my country who'd not consider those parked cars as an obstacle and you'd get a ticket, and probably you'd get your license suspended, too.

Where I am (Queensland, Australia), it's legal to cross the double white line to give cyclists clearance (1m for <60km, 1.5 for >60km). It's also legal to straddle the lane divider or drive on painted islands to do so.

I imagine the self-driving cars are going to have significant hurdles to adapt to different jurisdictions.

It would seem they need to test and adjust the car locally in each region before release. After they collect the data sets, they just need to finetune to the local dataset.

A self-driving car does not need to be able to handle every situation to still be useful. For example, if I had a self driving car that occasionally stopped and said "Please get in the drivers' seat, human driver required," it would still be a huge improvement over a normal car. And in the taxi fleet scenario, the cars could use alternate routes, or refuse a pickup in a place it can't get to and drop off as close as it can get.

By the time you are in the drivers seat, it will already be too late. A vehicle can't just stop and do nothing. A bad decision may be better than no decision at all.

It can in the scenarios listed above. If it sees the traffic lights blinking because a human is doing traffic control there, the car could safely stop and require a human to take over. If you're waiting behind someone who has stopped in the middle of the road to load or unload passengers, it's safe to just wait, or allow a human driver to figure out going around them.

That's not a self-driving car. A sometimes self-driving car is going to get people killed.

It has to be able to handle every situation safely, but doesn't need to be able to continue operating in every situation. If unexpected construction blocks the road, it needs to be able to stop and pull over, it can't go haywire and just accelerate right into a bulldozer. But it doesn't need to be able to navigate every situation. For example, a self-driving car which couldn't self-drive at night would still be hugely useful to me, I'd pay double for that feature.

The sooner they get the product in use by consumers the better. They've taken an overcautious approach and been too focused on partnering.

I really hope they'll deploy their own cars so that they have control over the platform and can iterate fast. There was a rumor that Apple was trying to buy McLaren. A similar deal might allow Google to move faster and get the product out there.

I think caution is warranted in dangerous situations like this. Accidents could be a major setback.

Human drivers caused 1.25 million fatalities in 2013[1]. When do we start blaming those deaths on our collective failure to deploy self driving cars?

[1] http://www.who.int/gho/road_safety/mortality/en/

Machines are always held to much higher standards than humans. It would be rational to switch to robocars as soon as they drive at least as safely as humans. In reality, even a small number of accidents caused by the AI will destroy all trust in the technology. Most people don't think in expected death rates.

I'm not sure, but certainly for a start, "when self-driving cars would reduce, not increase, that statistic."

Which isn't now.

Given that they are apparently getting out of the ISP business because of the startup costs involved, I'm wondering how they'll do getting into the car business, which must have even higher startup costs.

They're not getting out of the ISP business. Sure, it's been slowed down, but what did you expect to happen when national ISP's are slowing the deployments to a crawl by suing their local governments to try and block Google?

My theory is that they are getting out off Fiber / ISP business not because of costs, but rather the goal of Google Fiber has been met. Back in 2012 there was a real risk posed by ISPs, however with effective lobbying ensured Net-Neutrality becoming adopted as an official policy, and with next president likely being from the same party. There is no longer need for Google Fiber as it was in 2012. Also maybe there is another technology thats more lucrative which is on horizon and thus Google is recusing itself from making likely to fall investments in optical fiber network.

I don't agree with this at all. Google's projects don't always make money, but I don't think it ever does something without an intention to make money. The idea that Google Fiber existed solely to further a political cause seems far-fetched to me.

Also, net neutrality being in the Democratic Party platform is a far cry from it actually being law. The election isn't over yet and so we don't know who the next president will be and which parties will control which houses of Congress.

It was to further a political cause that allows them to continue to make money. In the end, it was still about the money.

It also made other ISPs start deploying ~gigabit/s service, which was another stated goal, no?

And for internet companies, users' access to the internet is a complementary product: the more and faster, the better.

Google has done this before e.g. they adopted a brilliant strategy, and drove up the auction prices in a spectrum option forcing FCC to adopt open rules. By threatening to build out a Fiber optic network, it forced the competition to respond.

https://www.cnet.com/news/spectrum-auction-google-wins-by-lo...

They aren't getting out of the ISP business; the article notes they are not expanding Fiber to new cities but are pursuing other avenues for providing broadband access.

Exactly. I mean, they provide internet access with balloons now: https://www.solveforx.com/loon/

> now

Project Loon started in 2011 (officially announced in 2013).

I'd assumed that they were going to license the technology to partners and that their current efforts were to create the reference implementation.

I'm still cheering for George Hotz in this one. Competition is good. I especially like scrappy types.

http://www.pcmag.com/news/347864/hacker-george-hotz-unveils-...

"The world is going to have cars that are sold to individuals and cars that are shared by individuals, and which one Alphabet does, we have our thinking on it"

As Tesla is going after the former ("affordable for everyone") it would make a lot of sense for Google to go after the shared market. It also feels more Google-y to me.

Makes sense. Google's technology is about ready for initial commercialization.

I had assumed it needed close collaboration with the maps/earth team and was mutually beneficial. If they spin it off I guess they could have some deal in place to preserve that aspect?

LAGUNA BEACH, Calif.—Google parent Alphabet Inc. is graduating its self-driving-car project from its research lab X into a stand-alone business, said X chief Astro Teller, a major step in the vehicles’ path to commercial operations.

The car group’s finances were separated from X on Jan. 1 this year, and now the team is completing a series of corporate and legal moves to become its own business, Mr. Teller said in an interview at the WSJDLive 2016 tech conference.

As its own stand-alone business under Alphabet, the car group would likely be expected to soon begin generating revenue, though not necessarily a profit at first. Mr. Teller declined to disclose the car project’s planned business model.

“The world is going to have cars that are sold to individuals and cars that are shared by individuals, and which one Alphabet does, we have our thinking on it,” he said. “But right now we’re very focused on safety.”

Mr. Teller said Alphabet will likely roll out its self-driving cars incrementally over the next several years as they improve with more time on the road. For example, the group could choose to launch the cars commercially in just a handful of cities with favorable roads and weather, before expanding to more challenging roadways and climates, he said.

He compared the process to teaching a teenager to drive, from parents first keeping their hands close to the wheel to letting the teenager drive alone only during daylight to eventually eliminating restrictions altogether.

“Any other company that’s acting rationally will probably do the same thing, which is finding constrained ways to roll out so that we’re learning safely,” he said.

Other companies building self-driving cars are deploying their technology gradually, though in many of those cases the auto firms are releasing semiautonomous technology that still requires drivers to be ready to take control.

Mr. Teller said Alphabet’s cars will be fully autonomous from the start. “You press a button and tell the car where you want to go,” he said.

Full Text^^

Reply on news.ycombinator.com