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

Self-Driving Teslas Are Fatally Rear-Ending Motorcyclists More Than Any Other

fuelarc.com

Post reply on HN
Print view View original

291–300 of 449 posts

Re: Self-Driving Teslas Are Fatally Rear-Ending Motorcyclists More Than Any Other

#291

Post by hatsix »

xhkkffbf wrote: post #232
xbmcuser wrote: post #22

The point I soured on Musk was when he ditched radar/lidar and tried to go with camera's alone. This made me realize he is not the genius he is made out to be but instead he is a fraud/charlatan and over the years his statements on different topics have only hardened that belief. Why the fuck would you want ai cars to be the same as humans you should want them to be many times better and radarr/lidar sonarr type tech…

I can see that cutting out the LIDAR could qualify as "cheap." And maybe he's being risky on betting on lowering the cost by simplifying the technology. But why make the jump to "fraud/charlatan"? Every system needs to be finite. We can invest in every bell and whistle. Furthermore, he's upfront about the decision. Fraud requires deception.

According to Musk, my car was supposed to be unsupervised driving by now. The shift to vision only has consumed all of their resources for the past several years, and my car has been left behind. There have been giant leaps in vision only, but it still isn't better than the vision+radar.

So, I was deceived. I didn't buy the car because of the deception, but I did buy FSD because of it.

Also, FSD disengaging when it gets sensor confusion should be considered criminal fraud. FSD should never disengage without a driver action.

Re: Self-Driving Teslas Are Fatally Rear-Ending Motorcyclists More Than Any Other

#292

Post by hwillis »

fsh wrote: post #20

This problem has been solved more than a decade ago by radar sensors (standard on many mid-range cars at the time). They detect imminent collisions with almost perfect accuracy and very little false positives. Having better sensor data is always going to beat trying to massage crappy data into something useful.

Backing this up: automotive radar uses a band at ~80 GHz. The wavelength is ~3.7 millimeters, which lets you get incredible resolution. Not quite as good as the TSA airport scanners that can count your moles through your shirt, but good enough to see anything bigger than a golf ball.

For a long, long time automotive radar was a pipe dream technology. Steering a phased array of antennas means delaying each antenna by 1/10,000s of a wave period. Dynamically steering means being able to adjust those timings![1] You're approaching picosecond timing, and doing that with 10s or 100s antennas. Reading that data stream is still beyond affordable technology. Sampling 100 antennas 10x per period at 16 bit precision is 160 terabytes per second, 100x more data than the best high speed cameras. Since the fourier transform is O(nlogn), that's tens of petaflops to transform. Hundreds of 5090s, fully maxed out, before running object recognition.

Obviously we cut some corners instead. Current techniques way underutilize the potential of 80 GHz. Processing power trickles down slowly and new methods are created unpredictably, but improvement is happening. IMO radar has the highest ceiling potential of any of the sensing methods, it's the cheapest, and it's the most resistant to interference from other vehicles. Lidar can't hop frequencies or do any of the things we do to multiplex radar.

[1]: In reality you don't scan left-right-up-down like that. You don't even use just an 80 GHz wave, or even just a chirp (a pulsing wave that oscillates between 77-80 GHz). You direct different beams in all different directions at the same time, and more importantly you listen from all different directions at the same time.

Re: Self-Driving Teslas Are Fatally Rear-Ending Motorcyclists More Than Any Other

#294

Post by smallmancontrov »

triplesec wrote: post #288
smallmancontrov wrote: post #254

Earlier quoted context omitted.

https://www.nhtsa.gov/laws-regulations/standing-general-orde... The numbers include all incidents where self-driving was active within 30 seconds of the crash, because of course they do. The meme that Tesla is allowed to simply bypass reporting on a technicality would be absurd if it weren't so voraciously spread by post-truth luddites. Look, if you want to dunk on Elon, I get it, but he does so many real shitty thin…

Guess what: Musk will probably get this rewuirement scrapped by his orange mate. https://www.reuters.com/business/autos-transportation/trump-...

Finally some good criticism in this godforsaken thread. Yes, Elon wants to do corruption and we should oppose him.

Re: Self-Driving Teslas Are Fatally Rear-Ending Motorcyclists More Than Any Other

#295

Post by triplesec »

ben_w wrote: post #273
david38 wrote: post #218

Earlier quoted context omitted.

A narrow case easily accounted for

Enumerating all the narrow edge cases has proven surprisingly difficult.

but train tracks is not that kind of edge case, tbh, it's so vanishingly situational

Re: Self-Driving Teslas Are Fatally Rear-Ending Motorcyclists More Than Any Other

#296

Post by Retric »

larvaetron wrote: post #164
Retric wrote: post #146

Earlier quoted context omitted.

Cars unlike aircraft don’t need to move forward to maintain safety. If the lidar normally has very good uptime but happens to break, you need to safely come to a complete stop and that’s basically it. Trusting vision systems for 30 seconds or even 30 minutes over the lifetime of a car is very different than trusting them for 30,000 hours. So for edge cases sure, add a “this is an emergency drive to a hospital without…

> Cars unlike aircraft don’t need to move forward to maintain safety. They do if they're crossing railroad tracks.

You can get out of a car on railroad tracks. However, “safely come to a complete stop” includes turns and freeways which are more serious concerns.

That’s why I said 30 seconds / 30 minutes not 3 seconds. The idea is to go to somewhere safe and pull off the road not just slam on the breaks and hope your not on a curve.

Re: Self-Driving Teslas Are Fatally Rear-Ending Motorcyclists More Than Any Other

#297

Post by vitus »

Spooky23 wrote: post #271
vitus wrote: post #200

Earlier quoted context omitted.

I think that's an oversimplification, and helps mainly in the case where one sensor is totally offline. If you have two sensors, one says everything's fine but the other says you're about to crash, which one do you trust? What if the one that says you're about to crash is feeding you bad data? And what if the resulting course correction leads to a different failure? I'd hope that we've learned these lessons from the…

The Elon magic is getting people to bikeshed bullshit like this and ignore the bigger issues. You don’t gouge out your ears because you hear something walking around at night that doesn’t appear to be accurate. As a human, your executive function makes judgements based on context and what you know. Your eyesight is degraded in the dark, so your brain pays more attention to unexpected sound. The argument for lidar, so…

What? I was responding to the specific claim that two independent sensors always make your failure rates go down. Not that any particular source of sensor data is strictly inferior.

I agree that optimizing for vehicle cost and shipping shoddy software at the expense of human lives is the wrong tradeoff, and having different sensors that provide more coverage of situations is generally preferable if you can afford it.

But suppose you have a radar sensor that's faulty, and so it periodically thinks that there's something right in front of it, and so it causes the car to slam on its brakes. That's likely going to cause an accident if you're traveling at highway speeds. Does that mean that we shouldn't use radar in any circumstances? Of course not. But sensors are not infallible, and we need to keep that in mind when designing these systems.

Re: Self-Driving Teslas Are Fatally Rear-Ending Motorcyclists More Than Any Other

#298

Post by bryanlarsen »

pwg wrote: post #277
bryanlarsen wrote: post #92

Prediction: over the next 4 years we're going to see lots of stories like this. Some of the stories will be fair, some won't. "Musk bad" stories get clicks. The next administration will be very anti-Tesla. Coincidentally, by that time self-driving will be considered mature enough to warrant proper regulation, rather than the experimental regulation status we have now. Combine the two, and the regulations will be writ…

> Combine the two, and the regulations will be written such that it excludes Tesla and includes Waymo. The solution seems easier, if only the regulators would pick up upon it. Under the current human driven auto regime, it is the human that is operating the machine who is liable for any and all accidents. For a self-driving car, that human driver is now a "passenger". The "operator" of the machine is the software wri…

Being both the owner and operator, Waymo already has full liability. It's a good proposal, but I don't think it's sufficient if your secondary goal is "screw Elon".

Nor is it sufficient to ensure that self driving is significantly safer than human drivers. I don't think the public wants "slightly safer than humans".

Re: Self-Driving Teslas Are Fatally Rear-Ending Motorcyclists More Than Any Other

#299

Post by breadwinner »

moralestapia wrote: post #87
xbmcuser wrote: post #22

The point I soured on Musk was when he ditched radar/lidar and tried to go with camera's alone. This made me realize he is not the genius he is made out to be but instead he is a fraud/charlatan and over the years his statements on different topics have only hardened that belief. Why the fuck would you want ai cars to be the same as humans you should want them to be many times better and radarr/lidar sonarr type tech…

>This made me realize he is not the genius he is made out to be but instead he is a fraud/charlatan and over the years his statements on different topics have only hardened that belief. That was Karpathy's decision [1] and, yes, I also have that perception of him. I know this is not going to be well received because he's one of HN's pet prodigies but, objectively, it was him. 1: https://www.forbes.com/sites/bradtempl…

From the article: While Elon Musk is best known for making statements on this, Karpathy was his go-to guy on backing up that reasoning.

I read that as Musk wanted this done and asked Karpathy to find a way.

Re: Self-Driving Teslas Are Fatally Rear-Ending Motorcyclists More Than Any Other

#300

Post by throwaway31131 »

potato3732842 wrote: post #285
throwaway31131 wrote: post #275

Earlier quoted context omitted.

> They generally can't detect stationary objects, have problems with reflections, most of them are VERY low resolution, and so on. Radar can absolutely detect a stationary object. The problem is not, "moving or not moving", it's "is the energy reflected back to the detector," as alluded to by your second qualification. So something that scatters or absorbs the transmitted energy is hard to measure with radar because…

Imagine if you could only see in a very narrow portion of the visible light spectrum, like only green or something. That's kind of how radar "sees" (I'm grossly over-simplifying here but point is it doesn't see the way we do). It's hard to tell something sticking off the back of a truck or a motorcycle behind a vehicle without false positive triggering off of other stuff and panic braking at dumb times, something ear…

I guess it boils down to this.

Earlier in the thread people were saying removing lidar was bad becuase multiple sensors types is good, presuming the camera stay either way, and one is not replacing camera with radar. I agree with this. It's usually trivially easy to corner case defeat one sensor type, as your example shows, regardless of sensor type. They all have one weakness or another.

That's why things like military systems have many sensor types. They really don't want to miss the incoming object so they measure it many different ways. Defeating many different sensor types is just way harder and therefore more unlikely to occur naturally.

And yes, control systems can absolutely reliably combine the input of many sensors. This has been true for decades.

Frankly I surprised more of these systems don't take advantage of sound. It's crazy cheap and society has been adding sound alerts to driving for a long time (sirens, car horns, train horns, etc.)

Post reply on HN