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

Tesla has a self-driving strategy other companies abandoned years ago

arstechnica.com

Post reply on HN
Print view View original

41–50 of 222 posts

Re: Tesla has a self-driving strategy other companies abandoned years ago

#41

Post by carimura »

munificent wrote: post #25
ForHackernews wrote: post #18

Earlier quoted context omitted.

Vision alone fails pretty frequently even with a human brain and 700 million years of evolution behind it. Count me as a skeptic that cameras + computer will produce a safe self-driving system.

> 700 million years of evolution behind it Sure, but only about 0.000014% of that time occurred while cars existed. It's not like we evolved to control big metal objects hurtling down artificially made flat surfaces.

...with devices that we're emotionally and physically addicted to flashing and making noises that are designed to ignite a flurry of activity in our reptilian brains sending the message PAY ATTENTION TO ME IMMEDIATELY!

Re: Tesla has a self-driving strategy other companies abandoned years ago

#44

Post by alkonaut »

syntaxing wrote: post #4

Not having LiDAR for anything over level 3 self-driving capabilities seems like a very bad idea...Computer vision right now just does not have the spatial awareness that you need for self driving capabilities. I wish Tesla would work on something similar to the Kinect v3 but for long range (above 20 m).

Regardless of whether cameras or lidar win out, just consider the fact that back up cameras already exist on tons of modern cars, and the solution to the environment issues of these cameras is one of:

a) fold it out when you put the car in reverse, so it's always ice-free

b) ignore this problem. It's covered in ice for months.

This is a simple problem in comparison, and neither of the above solutions would work for a camera used for autonomous driving.

Re: Tesla has a self-driving strategy other companies abandoned years ago

#45

Post by matthewdgreen »

pjkundert wrote: post #23
whttheuuu wrote: post #8

ITT: people who think they know better than the company that has been working on this stuff day in and day out for years lol.

... working on this stuff day and night ... —- in California . I’ve been driving for 40+ years, millions of km, 1/2 during winter. Now, Elon is a Saskatchewan boy, so he should know better. Perhaps he’s forgotten. Winter driving is at best 1/2 vision dependent, often much less. Many times, you have to actively ignore your vision (everything you see is “moving” sideways). Quite often, you’re modulating your throttle t…

The vision problems are real. But the ability to handle ice/snow-induced wheel spin and slew seems like exactly the kind of problem that can be handled with sensitive electronic traction control, perhaps augmented with accelerometers. That specific problem doesn't seem intrinsically more complicated than, say, keeping a drone level on a breezy day -- which is something machines already do much more reliably than humans can. (New England driver here, so I do have some personal experience.)

Re: Tesla has a self-driving strategy other companies abandoned years ago

#46

Post by alkonaut »

ableal wrote: post #3

"Self-driving cars also benefit from lidar sensors, and the best ones cost thousands—if not tens of thousands—of dollars each. That's too expensive for an upgrade to a customer-owned vehicle. But the economics are more viable for a driverless taxi service, since the self-driving system replaces an expensive human taxi driver." That's probably the crucial point. For now, lidar is needed for safe operation, and too exp…

Commercial (not even taxi, but transport) will be the first to be autonomous because of the larger savings possible. That we'll see useful autonomy for a $50k car driven by individuals, rather than in a $1M transport vehicle operated by a huge logistics company isn't likely.

Re: Tesla has a self-driving strategy other companies abandoned years ago

#47

Post by leesec »

ForHackernews wrote: post #18
cbames89 wrote: post #5

This is a (perhaps short-term) failure of first principles reasoning. Elon's known to favor thinking about things from first principles, and there's no theoretical reason that vision can't work. However, technical limitations might take a while to overcome.

Vision alone fails pretty frequently even with a human brain and 700 million years of evolution behind it. Count me as a skeptic that cameras + computer will produce a safe self-driving system.

Really doesn't though. Humans are pretty good at driving, and 90%+ of accidents come from people being drunk/distracted/tired etc. We don't crash because we miss categorize a truck as a plastic bag or w/e.

Re: Tesla has a self-driving strategy other companies abandoned years ago

#48

Post by RivieraKid »

cbames89 wrote: post #5

This is a (perhaps short-term) failure of first principles reasoning. Elon's known to favor thinking about things from first principles, and there's no theoretical reason that vision can't work. However, technical limitations might take a while to overcome.

People should strive for correct reasoning. First principles reasoning, as I interpret that term, is just an approach that helps to achieve correctness by overcoming ingrained assumptions.

If Elon really thought "in theory, camera-only full self driving is possible, therefore I should invest resources in that approach" than he's dumber than I thought. He's done it because Lidar was not an option - and the claim that FSD is coming soon was an intentional lie.

Re: Tesla has a self-driving strategy other companies abandoned years ago

#49

Post by Animats »

Waymo keeps plugging away. Each year, for the last few years, the number of miles between disconnects they report to the CA DMV has doubled. Three more doublings, and that number will be bigger than average miles between accidents for human drivers. Then they can ship a product.

Nobody else is even close.

This problem is slowly being solved, by normal engineering practices. The "fake it til you make it" players are being left behind. Uber has been shown to be totally incompetent. Tesla really just has a good lane follower and an mediocre car follower. Apple has been trying to hand-wave by talking about "significant disconnects" vs. the all disconnects DMV requires them to report.

The LIDAR industry is struggling, but there is progress. Quanergy seems to have been mostly hype.[1] Continental, the big European auto parts maker, bought Advanced Scientific Concepts, which makes and sells a good but expensive flash LIDAR used in DoD and space applications. They packaged it up for automotive use, and are waiting for the self driving industry to catch up. That technology uses exotic indium-gallium-arsenide sensor ICs, which are expensive in small quantities but would probably be affordable if they could sell a few million a year.

This looks like a problem that's being solved. Just not fast enough for startups used to quick payoffs in software.

[1] https://www.bloomberg.com/news/features/2018-08-13/how-a-bil...

Re: Tesla has a self-driving strategy other companies abandoned years ago

#50

Post by andrewtbham »

martythemaniak wrote: post #30

There's a few things to keep in mind when discussing self-driving tech. 1) Self driving tech doesn't exist today. It simply does not. There's lots of people working on it and they are using different strategies, but it is not clear how long it will take, which strategy is technically superior, which strategy is more economical, etc etc. There's a lot of very strong opinions floating around (LIDAR is essential! No, vi…

I completely agree with these points....

Also... if you bought full self driving you're going to get the new v3 hardware retrofit... that's what you paid for and it's going to be better... probably a lot better.

(edit: removed link... just google it)

Post reply on HN