Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Toyota Research Institute head says full autonomous driving is “not even close”

techcrunch.com

Post reply on HN
Print view View original

371–380 of 443 posts

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#371

Post by fastball »

kpil wrote: post #233
kajecounterhack wrote: post #209

Earlier quoted context omitted.

I don't know by what metrics you're making these comparisons. Audio recognition and self-driving vehicles are apples and oranges. Autonomous driving is difficult primarily because of a changing world and low tolerance for mistakes. The state space of situations you're trying to map is just large. It will take time. But it's not even close to impossible. Audio recognition is difficult for a different reason: language…

I think that audio recognition is an order of magnitude less complicated than fully autonomous driving. The world is fully contextless outside the road to an not-even-at-insect-level brain driving two tons of metal, and anything can happen. Voice recognition is getting just barely tolerable. If you have tried to do something important using voice recognition you'd probably go bat shit crazy after a while. If it had t…

  I think
Wonderful.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#372

Post by fastball »

Piskvorrr wrote: post #140
Bartweiss wrote: post #80

Earlier quoted context omitted.

One point would be that both planes and trains are professionally operated already. That means three of the biggest incentives to automate are missing. First, they have few human-error crashes; they're both significantly safer than automobiles, so unless automation can do better-than-human disaster recovery (not likely yet) there's minimal safety advantage. You have to outperform well-trained humans instead of random…

Good points. Automated train control maps are not even 2D: they're directed graphs, precisely mapped to sub-foot precision, with all permissible edges fully checked in advance, at design and at mapping. Far easier to navigate than a road network, which can be simplified and approximated to a directed graph, but 1. at the cost of some precision, and 2. orders of magnitude more likely to change unnoticed.

  Unions.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#373

Post by throwaway346323 »

TYPE_FASTER wrote: post #311
Animats wrote: post #173

Toyota got into autonomous driving late. Then they hired Gil Pratt from MIT. Pratt used to run the MIT Leg Lab, which Raibert ran before Boston Dynamics, but Pratt didn't do much there. It's clear that autonomous driving can be made to work, because Google/Waymo is doing it. It's hard and expensive and it takes a lot of sensors. It also takes extensive testing. Waymo drives 25,000 autonomous miles a week. Volvo has l…

> Toyota got into autonomous driving late. Then they hired Gil Pratt from MIT. Pratt used to run the MIT Leg Lab, which Raibert ran before Boston Dynamics, but Pratt didn't do much there. They've hired other folks, too, who have real-world autonomous vehicle experience. Google/Waymo are definitely leading the pack, but I don't think we'll see Level 5 in a production vehicle for at least five years, probably closer to…

They're also heavily invested in PFN which makes Chainer (and is apparently helping them on the self-driving effort in Tokyo ).

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#374

Post by ghaff »

euyyn wrote: post #189
ghaff wrote: post #149

Earlier quoted context omitted.

>Driver salary is 52% of the cost of a taxi. If you drop the utilization, the cost of the driver as a percentage probably goes up a bit. But here's the thing about those numbers. Yes, driver salary is a big chunk of taxi cost but flip things around. It's only about half the cost. In other words, there's this widespread assumption that autonomy totally changes the nature of automobiles and automobile ownership even th…

Getting a 50% discount on taxi rides would make me never consider any other option when I need a taxi.

It depends on the circumstances. On the other hand, for me, if taxi rides were half the price I doubt it would affect my ridership at all over the course of a year. MAYBE for one or two marginal car rentals.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#375

Post by Chaebixi »

Animats wrote: post #347
Swizec wrote: post #334

Earlier quoted context omitted.

> That's the approach we know won't work. That was Tesla's first-round autopilot, the self-crashing car. I think OP was talking about "crash avoidance" systems. Not systems that claim to drive the car but not really. You 're always driving, the car is just silently observing. If the car sees you're about to crash, it steps in and takes over. The opposite of Tesla's approach where you play the role of observer taking…

There are many automotive "auto-brake" systems, usually controlled by a radar. Those are useful. They're usually programmed to act only when a collision is inevitable. They prevent some slow-speed collisions, and reduce the damage in some faster ones. What they don't do is take over steering. A last-second takeover of steering would make things worse some of the time.

> What they don't do is take over steering. A last-second takeover of steering would make things worse some of the time.

I have a Honda. While its lane-keeping system won't steer during a crash, it definitely does steer to keep you centered in the lane (when it's able). That could help avoid all kinds of crashes due to inattentive lane-drifting.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#376

Post by bb611 »

euyyn wrote: post #185
SEJeff wrote: post #54

Earlier quoted context omitted.

Planes are mostly automated. The pilots are involved for takeoff and landing or emergencies, but little else. All modern Airbus or Boeing planes are mostly flown with autopilot. The human component is still involved primarily for the "what if it breaks" case where the software isn't there today. A computer would likely have never considered putting a plane into the Hudson river, but that pilot saved a lot of people b…

There's also automated take off. Fighter pilots taking off carriers have to grab a couple of handlebars in the cockpit, whose purpose is only that the pilot doesn't instinctively intervene during take off.

There is no automation in takeoff for current USN carrier based aircraft. The F-18 is the most automated, and the process is this:

1. Pilot sets takeoff conditions (throttles, flaps)

2. Pilot signals deck crew

3. Catapult accelerates the plane off the deck

4. The plane's fly by wire control system pitches the plane for optimal Angle of Attack for the wings (IIRC - 8.1* without flaps).

Once the aggressive acceleration of the catapult is complete, the pilot takes control again and begins flying - this is less than 1 second after catapult release.

Here's an example catapult launch from a USN F-18 pilot: https://www.youtube.com/watch?v=Nj9D1Ls-_JM

If you look at the stick directly in between the pilot's legs, you'll see him put his right hand on it at ~4.5 seconds. His left hand is on the throttles.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#377

Post by will_hughes »

memnips wrote: post #368
Skunkleton wrote: post #277

Earlier quoted context omitted.

Perhaps a portion of Tesla's problem was with marketing? Autopilot gives very different cues than something like LKAS (lane keep assist).

And it's also called "Autopilot" which sets an extremely different tone and expectation than other driver-assistance/safety technologies.

> And it's also called "Autopilot" which sets an extremely different tone and expectation than other driver-assistance/safety technologies.

Only to those not familiar with autopilot systems in aircraft.

It's like the use of "hacker" - here on this site it still (probably) has most of it's original meaning. To the rest of the world it means a guy in a ski-mask with a russian/chinese accent who's after your bank account.

Unfortunately most of the world thinks pilots in commercial aircraft turn on the autopilot and can then go for a sleep, so when they see "Autopilot" on a Tesla, they think they can do the same.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#378

Post by smileysteve »

mandevil wrote: post #315
ageofwant wrote: post #179

Earlier quoted context omitted.

Intercontinental planes are mostly autonomous. Trains are autonomous. Massive mine dump trucks are fully autonomous. These things are in use today and has been for almost a decade. They need not be perfect to be useful, they only need to be better than the average human driver. Toyota is way behind in the game. What do you expect them to say about the competition?

Planes are a good argument for the opposite. Do you ever wonder why there are two pilots on every commercial airliner? It's not because airlines like paying so many more pilots- notice how fast flight engineer/navigator positions disappeared as flight management technology and GPS made them unnecessary. But still two pilots are there on every commercial flight, why? Simple answer: the airline industry has learned, th…

Because you reference AF447; Autopilot systems also added additional sensors (satellite assisted speed) for the system to have redundancies.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#379

Post by averagewall »

jccooper wrote: post #161
ZeroGravitas wrote: post #95

Earlier quoted context omitted.

"This became painfully obvious with the article floating around where a car misread a stop sign with a small amount of marking as a 45MPH speed limit sign. It's pretty clear that if a car is mistaking a red sign for a white sign, we have a very long time to go before the car can safely drive itself without a human observer to intervene when needed." I think you misread that article. Someone built their own toy implem…

While they did use their own simple DNN classifier (because they couldn't find one), it was treated as a black box for purposes of the attack. A more robust classifier (probably using multiple methods) would be more resistant to attack, but the attack would still work, just the changes would be more obvious. At some point you hit "human-recognizable" but it's unclear where that point is.

As long as you have to intentionally modify the sign to cause confusion, as opposed to doing random vandalism, I don't think this is a problem at all. It's like a person spray painting extra loops to change a 30 into an 80 or sticking a home-made sign over a real one so it's indistinguishable to most drivers. People aren't going to do that any more than they already do and it'll still be just as illegal.

Re: Toyota Research Institute head says full autonomous driving is “not even close”

#380

Post by averagewall »

cmurf wrote: post #279
cryptoz wrote: post #25

Earlier quoted context omitted.

30,000 people in the USA alone are killed each year because of human driving error. Humans are not getting any better at driving. Car safety mechanisms have reduced road death rates significantly in the last 50 years, but still - 30,000 deaths/year in the USA alone due to human driver error. Cars will be automated because cars have to be automated.

About the same number of people die of influenza each year. We've been dying in these numbers for a long time, and it really doesn't matter at a macro scale. It's fine to try and do better, there are some positives for trying, but this is more of a want, not a need.

Those are mostly the elderly and babies who have a high risk of dieing anyway. Elderly people especially are certain to die of something soon. If it's not the flu, it'll be a cold or an infection or whatever.
Post reply on HN