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

Tesla in autopilot mode crashes into parked Laguna Beach police cruiser

latimes.com

Post reply on HN
Print view View original

331–340 of 488 posts

Re: Tesla in autopilot mode crashes into parked Laguna Beach police cruiser

#331

Post by cherioo »

DannyBee wrote: post #307
mikeash wrote: post #169

Earlier quoted context omitted.

I'm extremely skeptical that there have been zero TACC-involved crashes with other cars. Has the NTSB or other organization actually put out a statement to that effect somewhere, or are you just basing this on an inability to locate any when searching?

Again, please provide any data that says otherwise. At all. Any! So far the data in this conversation has been incredibly one sided, with me providing data, and you providing skepticism. These systems exist in a lot of cars, yet nothing until Tesla suggests any issues anywhere (i'm ignoring things like news reports which are fairly biased) Therefore, the extraordinary claim is not that Tesla is worse, it's that Tesla…

A quick search gives me this http://www.thetruthaboutcars.com/2016/08/automatic-emergency.... In particular "At 45 mph, the crash-avoidance systems slowed the test vehicle by 74 percent, on average, and prevented collisions 40 percent of the time.", i.e. it fails 60 percent of the time.

Do you really believe other cars' automatic emergency braking works 100 percent of the time with perfect reliability? They don't show up because it's non-news.

Re: Tesla in autopilot mode crashes into parked Laguna Beach police cruiser

#332

Post by joshuakcockrell »

Animats wrote: post #276

Has Tesla blamed the driver yet? They might get away with that here. The road was not divided and unsuitable for their lane-keeping system. The pavement markings are unusual.[1] Of course, as usual, their obstacle detection failed to detect a stationary obstacle that didn't look like the rear end of a car in the same lane. We need a minimum standard for a vehicle which takes automated control of steering and braking.…

Here's the exact spot on that road that matches the picture in the article https://goo.gl/maps/pj3X4gAuG672

Re: Tesla in autopilot mode crashes into parked Laguna Beach police cruiser

#333

Post by simion314 »

wooter wrote: post #219
beamatronic wrote: post #180

Earlier quoted context omitted.

I feel like I-5 between SF and LA is probably one of the safer places to use it. Any place else, maybe not so much.

typical fear-based, anti-freedom, anti-progress response in these comments. The NHTSA found a 40% crash rate reduction from Tesla's autopilot. So yea "blame Elon Musk"

This is false, stop this Tesla propaganda with false facts.

Re: Tesla in autopilot mode crashes into parked Laguna Beach police cruiser

#334

Post by hndamien »

hannasanarion wrote: post #296
hndamien wrote: post #55

Earlier quoted context omitted.

Auto pilot != Self driving. People love to conflate these two so much that the people saying it rely on everybody else conflating it to justify their ignorance. This is just like Fox "some people say" - Fox people say! Or Javascript Bitcoin miners - Monero is not Bitcoin! It's snowing so there is no global warming - It is "climate change" but the entire globe is warming! This hurricane is just weather, not climate ch…

You're like Lewis Carroll's Humpty Dumpty > what I say is always what I mean, and if you misunderstand, it's your fault. Linguistic pragmatics 101: communication is a two way street.

So Wikipedia's consensus that I linked to is wrong?

Re: Tesla in autopilot mode crashes into parked Laguna Beach police cruiser

#335

Post by forapurpose »

hfdgiutdryg wrote: post #96
njarboe wrote: post #74

Earlier quoted context omitted.

But not any worse than the term cruise control really.

You're getting downvoted, but I agree. A commercial pilot doesn't flip on autopilot and start browsing the internet on a phone or take a nap. And that's at 30k feet where there's little traffic to worry about. It seems like a misunderstanding of what the technology does, by consumers and the media. I'm no expert on Tesla's marketing history, but I'm of the opinion that these crashes are due to drivers misunderstandin…

> A commercial pilot doesn't flip on autopilot and start browsing the internet on a phone or take a nap

Commercial pilots are trained professionals; Tesla drivers are not.

Re: Tesla in autopilot mode crashes into parked Laguna Beach police cruiser

#336

Post by simion314 »

wooter wrote: post #237
joejerryronnie wrote: post #214

Earlier quoted context omitted.

"Move fast and break things" was never meant for actual real world physical things (or laws). It was a mantra that encouraged breaking sacred cows and assumptions but has been bastardized and vilified by the anti-big tech crowd. Obviously you can be a little more lax on your production quality if you're building inane web apps rather than rocket ships or medical devices.

not to mention that autopilot has demonstrably reduced the crash rate by 40% so it really is just vilification from a desire for retribution without consideration of actual impact

False, read again the facts and stop spreading wrong conclusions or facts, btw the study you are thinking off is not open so it was not verified, I think there is an action in justice to open the data.

Re: Tesla in autopilot mode crashes into parked Laguna Beach police cruiser

#337

Post by cherioo »

People always lament about how Tesla tries to mislead customer's about its capability, but is there actually data that shows this is truly the case among Tesla owners? How many times does "attention needed" beep is needed before customer can be considered reasonably informed that it is not fully self-driving?

Even among non-Tesla cars, whose manufacturers don't try to "mislead", 71 percent of people believe automatic emergency braking can avoid all crashes [0]. Is this percentage higher or lower for Tesla, and is that difference warranted?

[0] https://newsroom.aaa.com/2016/08/hit-brakes-not-self-braking...

[1] Quote from above "When traveling at 45 mph and approaching a static vehicle, a scenario designed to push systems beyond the stated limitations, the systems designed to prevent crashes reduced speeds by 74 percent overall and avoided crashes in 40 percent of scenarios. In contrast, systems designed to lessen crash severity were only able to reduce vehicle speed by 9 percent overall."

Re: Tesla in autopilot mode crashes into parked Laguna Beach police cruiser

#338

Post by cr4zy »

One way to avoid these types of crashes IMO is anomaly detection. It's quite simple to do anomaly detection in pixels using modern deep pixel prediction nets like PredNet. In my experiments you get a few seconds lead time on something like a car cutting you off (the car starts to head out of the lane before actually crossing it for example). This allows alerting the driver, and with a full windshield HUD you could even highlight the anomalous pixels on the windshield. The nice thing about this is that it can be trained in an unsupervised manner on all the available data. Some important details are to find anomalies in object bounding boxes, using something like Tensorflow's object detection pretrained net. Otherwise buildings with lots of striations would light up the anomaly detector. Also, you should detect anomalies in a human colorspace like CIELAB so that white cars (#fff) are not artificially weighted as more anomalous.

Finally, you could use this as input to a planner like Model Predictive Control where a higher cost is incurred for approaching anomalous objects.

Re: Tesla in autopilot mode crashes into parked Laguna Beach police cruiser

#339

Post by maxxxxx »

megy wrote: post #218
confiscate wrote: post #91

I don't get it, why is it so hard to detect stationary objects in front of you? wouldn't that be the "hello world" of any self-driving tech? Every time I ask this question, the response is "radar as a technology has too much noise, not accurate enough to detect stationary objects". But it DOES detect stationary objects. Otherwise no one would even turn on Tesla AP at all. Why is it so hard to detect stationary object…

> I don't get it, why is it so hard to detect stationary objects in front of you? This could only be said by someone who has never worked on the problem. I don't get it, why is programming so hard, you are just telling the computer exactly what to do?

It seems a legitimate question. I have never worked on this problem either but I would have thought that detecting a stationary object right in front of the car should be one of the basic and simpler functions of a self driving system.

Obviously this doesn't seem to be the case but maybe you can enlighten us why this is hard?

Re: Tesla in autopilot mode crashes into parked Laguna Beach police cruiser

#340

Post by dang »

hndamien wrote: post #334
hannasanarion wrote: post #296

Earlier quoted context omitted.

You're like Lewis Carroll's Humpty Dumpty > what I say is always what I mean, and if you misunderstand, it's your fault. Linguistic pragmatics 101: communication is a two way street.

So Wikipedia's consensus that I linked to is wrong?

You've unfortunately contributed the most to this flamewar. We don't want those here. Please stop.

https://news.ycombinator.com/newsguidelines.html

Post reply on HN