Hacker News Reader

> Tesla's "autopilot" is just automatic lane keeping and automatic cruise control.

And lane changing. But yes, they never claimed the current system could do any more than that.

> Mercedes and BMW have been shipping that for years.

Yet, their systems on their 2017 models still isn't as good as Tesla's current system. [1]

> The other car companies have hardware to check if the driver's hands are on the wheel. Tesla not only didn't do that…

Tesla's system will indeed periodically prompt you to hold the steering wheel, and will disengage if the driver does not react to several warnings.

> … they hyped those basic functions as being an "autopilot", as if they had something comparable to what Google has

No. What Google has is called autonomous driving. You've completely misunderstood the meaning of the word "autopilot". From Wikipedia:

> An autopilot is a system used to control the trajectory of a vehicle without constant 'hands-on' control by a human operator being required. Autopilots do not replace a human operator, but assist them in controlling the vehicle, allowing them to focus on broader aspects of operation, such as monitoring the trajectory, weather and systems.

And that's exactly what Tesla's Autopilot does: assist the driver in controlling the vehicle, but does not replace a human operator.

[1] http://mashable.com/2016/06/16/2017-mercedes-benz-e-class-fi...

"Autopilot" is understood to mean "the plane flies itself" by the general public. Here's one human pilot who's ticked off by it:

http://www.askthepilot.com/questionanswers/automation-myths/

> "Autopilot" is understood to mean "the plane flies itself" by the general public.

Just because people mistakenly believe something doesn't make it right. The general public once thought the Earth was flat, and how did that turn out?

Unlike the shape of the Earth, there is no truth of the matter regarding what a word means except common usage. In certain cases a word might have a technical definition beyond the usual consensus but 'autopilot' isn't one of those words and communications to the public in, say, an advertisement have to use the common meaning of words.

>Just because people mistakenly believe something doesn't make it right.

Tesla promotes it as autopilot. Words matter.

If Tesla itself selectively denies that the common understanding is the meaning they intend, then they appear to be trying to have it both ways: cashing in on the PR and cachet that the common meaning confers, while disclaiming the responsibility that goes along with owners actually believing it.

You're right, Tesla mistakenly believed they could cash in on calling their system Autopilot™ and somehow sidestep the connotations with the name and the damage pushing a broken implementation on consumers is causing.

The unfortunate truth is, if I gave my mom a car with the autopilot feature, she'd end up eventually trusting it and not care about the implementation details that make Tesla's system not an automated driving feature. That's scary and dangerous to market to consumers.

On a sailboat "autopilot" generally means "keep this heading" or "keep this point of sail." I'm pretty sure in aviation it means "take me to this way point and elevation" or a sequence of such maneuvers. A plane flies itself or a boat navigates itself only so long as nothing unexpected occurs. In either case, a pilot must be standing by, maintaining awareness, ready to assume control and handle anything that may come up. I'm not sure what people think this means in terms of driving a car but I think it's early enough that Tesla can help define what this means.

Autopilot systems in aviation have become so complex and diverse that there is no single catch-all sentence to describe them. I guess you could come close with the phrase "a system designed to keep and guide the state of the aircraft within and through certain set constrains".

Those constrains could be constant like a certain heading or a direction (not the same thing by the way), a series of vectors or a single vector (like a flight-path or an approach to a runway) or constraints that depend on dynamic variables like controlling the altitude and speed according to the weight of the aircraft or automatically avoiding traffic.

Of course the pilot must be standing by because a) autopilots do not keep track of everything (and aviation control interfaces are quite complex) and b) sensors can and do malfunction.

This is not a setback for automatic driving. This is just a Tesla problem.

Let's all hope that's how this is remembered 10 years from now (or whenever) when automatic driving is actually ready.

It's not a given that people will remember this as a Tesla-only problem. Instead it might be remembered as that time when autopilot got people killed.

And when it's actually ready... what Tesla is doing today, might delay or slow the adoption of real autopilot. And that would be a real setback for the real thing.

there will be deaths and accidents from the computer controlled cars. that is pretty much inevitable. the question is where on the x of y miles driven per accident/fatality does it make sense to accept that computers and algorithms are fallible just like humans but in aggregate the system works better than just humans doing all the manual control inputs. it's the first of its kind and we only have one datapoint w.r.t. the deadliness of autopilot used without oversight but it seems to indicate that the autopilot is safer overall in aggregate.

No, it's not clear at all that autopilot is safer. Tesla has stated that it's safer by comparing the miles autopilot has driven vs the overall average for a human.

But that's an unfair comparison.. because humans drive in all conditions, while autopilot is more likely to be used on highways (and in good weather). Even human drivers are 3x safer on highways. So until Tesla is willing to compare like with like, we do not know how it compares.

It's also too early to waive off the damage that Tesla may cause. It's been 6 months since they rolled this thing out.

Over the next few years, they're going to give 300,000 cars to consumers with this enabled.

The few incidents that have been reported may just be the tip of the iceberg. We'll see how bad it gets when you have a few hundred thousand people driving around with what they think is autonomous driving.

I think a more apt comparison is miles driven while distracted. the guy was watching a movie and completely not paying attention. How many accidents were and are caused by people texting and driving per x number of miles? There were apparently enough that they passed laws against distracted driving in many states in america.

This is a beta product that explicitly says not to use it without paying attention. there are alarms that are supposed to go off and the vehicle is supposed to slow down to a halt if you do not have your hands on the steering wheel.

I see it like the beginning of the airline industry. At first they weren't as comfortable and safe as they are now. any time they have an airline disaster they do a thorough investigation and if it is warranted they will issue new mandates that airlines need to follow. over time, every accident makes air travel safer from those mistakes/edge cases/etc.

In ten years, when there is much more reliable autonomous driving, I think this will simply be remembered as a failed and misguided earlier attempt just like people trying to fly by strapping wings to their backs and flapping.

> The other big Tesla problem is a pure technical design error - the bumper-mounted radar is blind at windshield height. This is the immediate cause of the fatal accident involving under-running the semitrailer. It's also the cause of the parking accident with the truck with overhanging load sticking out the back.[2] Tesla's implementation has deadly blind spots. They need a radar at windshield-top height. It's quite likely, now that the NHTSA is looking into this crash, that Tesla will have to do a recall and retrofit one. (Along with hands-on-wheel sensors.)

This is just completely false. We've been putting them at bumper height for 10 years because that's where the system ends up working best with the greatest number of targets to see. NHTSA has been regulating these sensors for years and designing the tests for them; they are not new to this. Tesla is not going to have to do a recall on these, and spare me the idea that they are going to "retrofit a sensor at windshield height." This is just mindless speculation you made up.

This is just completely false. We've been putting them at bumper height for 10 years because that's where the system ends up working best with the greatest number of targets to see

Who's "we"?

Based on Josh's past comments he appeares to work for tesla on their radar systems.

I do not work at Tesla or for Tesla, just have experience with the radar systems.

Sorry for the mistake.

Nope, sorry on my part for not wording things better.

> because [the bumper is] where the system ends up working best with the greatest number of targets to see

Do you have a source for this? It was my understanding Google went with their roof-mounted approach because of the blind-spot issue. (They also went with Lidar over cameras because of the white-truck-in-bright-daylight problem.)

I worked on the radar systems personally and we tested them with the Lidar hardware that Google uses. When the systems are raised they become more difficult to engineer (Caterpillar uses similar sensors on its large trucks, which are raised). Ultimately we solve this with more capable radar hardware in the bumper, not multiple mounting points, though Volvo's innovation in this area is significant and commendable.

Volvo has placed radars at the top center of the windshield. They had a blind spot at bumper level, so later Volvo systems added a second radar down there. And a LIDAR. [1] One radar at bumper level is just enough to keep from rear-ending the car in front. That's not enough for a hands-off "autopilot". You need at least two, or beam steering in elevation, like the Fujitsu unit.[2] (Although that has only ±9° of elevation, not enough at bumper level to eliminate the need for a higher sensor.)

[1] http://www.pcworld.com/article/2025386/volvo-v40-watches-the... [2] http://www.fujitsu-ten.com/business/technicaljournal/pdf/38-...

I had not seen some of that Volvo tech, thanks for the link. Ultimately the radar portion is going to be more three dimensional tech still in the bumper. It's a struggle financially to get a single one of these sensors on a commodity vehicle, so the tech is focused on how to do more with that. Still, what Volvo is doing with the technology available is very interesting transitional technology.

I don't know why you would confidently write 8 paragraphs on something you clearly don't know much about. Here are a few corrections:

- tesla does have hands on wheel sensors

- tesla does not use stereo cameras (autopilot uses a single forward facing camera combined with radar and sonar)

- the second crash 'from yesterday' you linked is from May

Tesla doesn't seem to actually sense hands on wheel. They sense torque on the steering wheel.[1] Mercedes seems to have capacitative touch sensing, although customers complaining about it aren't quite sure.[2]

Tesla doesn't take the absence of hands too seriously. "I'm driving down the highway at 55, and I'm not looking, and I have no hands on the wheel".[3] Hands off driving in New York City traffic.[4] "I managed to go 50 miles without a single nag"[5]

Apparently Tesla does have only one camera on currently shipping products. They must be trying to do stereo from motion. There are rumors of a revision to the Model S with multiple cameras.[5]

Yes, the van crash was in May, not yesterday.

[1] https://forums.teslamotors.com/forum/forums/how-get-rid-hold... [2] http://mbworld.org/forums/new-s-class-w222/626411-steering-w... [2] https://www.youtube.com/watch?v=LS7pAC0OmLI [3] https://www.youtube.com/watch?v=M4BGlQoASyo [4] https://forums.teslamotors.com/forum/forums/nag-hands-wheel-... [5] http://cleantechnica.com/2016/05/04/new-tesla-model-s-2nd-tr...

> Mercedes seems to have capacitative touch sensing

This is not correct.

https://www.youtube.com/watch?v=Kv9JYqhFV-M

> autopilot uses a single forward facing camera combined with radar and sonar

It is actually camera-only per comment of Mobileye's CEO.

I think that's what Mobileye sell to Tesla. Per comments on Twitter[1] (and the fact that they fit a radar) the car's combining output from the Mobileye unit with the other sensors when decision making.

1. https://twitter.com/elonmusk/status/748625979271045121

> Here's another Tesla crash, from yesterday.[1] This is a Tesla rear-ending the corner of a stalled van on a highway.

> [1] https://www.youtube.com/watch?v=qQkx-4pFjus

minor nitpick but this video is from May, this didn't happen yesterday, did you mean to link another video?

You're right; that's from a few weeks ago. It got some publicity yesterday as if it were newer.

The Tesla owner says in the description "Yes, I could have reacted sooner, but when the car slows down correctly 1,000 times, you trust it to do it the next time to. My bad.."

And this is the same problem mentioned elsewhere in this thread. The adaptive cruise being 99.99% effective at stopping trains the driver that it will always work. Especially because Tesla doesn't force the driver to touch the steering wheel, they seem especially susceptible to this issue - and drivers do things like this: https://www.youtube.com/watch?v=5fDbfy0409U&t=0s

This is not an issue until it is.

It seems like the solution is to have the software train the driver to take control in unusual situations.

One way to do that might be that if the car successfully avoids an obstacle on its own without the driver reacting at all, after the obstacle is passed the car pulls over and stops.

> This is the immediate cause of the fatal accident involving under-running the semitrailer.

Rear ending a semi is one of the most fatal types of car crashes. People should definitely not trust automatic cruise control when driving behind a trailer. And regular drivers should give themselves some extra space when driving behind a semi.

https://youtu.be/bT3G-kcKN70?t=1m9s

I'm convinced, why don't we have stronger underride standards?

The trucking industry is powerful and will work to dodge anything that increases their costs like this. To be fair, the same is true of other industry representation.

This wasn't a rear-end. This was a side hit after being cut off by a semitrailer entering a four-lane but non-limited-access road from a side road.

> It's quite likely, now that the NHTSA is looking into this crash, that Tesla will have to do a recall and retrofit one. (Along with hands-on-wheel sensors.)

Will they? Considering that those are driver-assistance systems it's always the human in charge, no?

The driver would also have died if he had just enabled "dumb" cruise control and noticed the white trailer.

Of course more advanced automation systems should do better and tesla should improve, but from a technical perspective it's still just an assistance system.

Although I have no idea what the NHTSA will do, it's clear that people are treating Tesla's autopilot as more than a driver-assistance system.

Hopefully in the future Tesla will add some additional checks to make sure the driver is actually using it as an assistance system instead of an auto-pilot. And when their auto-pilot actually becomes an auto-pilot they can remove those checks.

>And when their auto-pilot actually becomes an auto-pilot they can remove those checks.

And then they can call it autopilot again...

I wouldn't be surprised if Tesla get slapped for marketing it as an autonomous driving system. Musk has publicly said their autopilot is twice as safe as a human driver. That is misleading at best considering it should never be used without a human driver.

There seems to be third problem: Tesla refuses to take responsibility for its product's shortcomings and blames the victim instead.

Well Tesla's first fault and major one is their naming of the feature.

The naming and marketing of it. Will they admit fault and change?

Perhaps drivers overestimate the capabilities of autopilots in aircraft? A pilot still has to practise see and avoid when using an autopilot.

> Perhaps drivers overestimate the capabilities of autopilots in aircraft?

Of course they do, which is why it's such brilliant and irresponsible marketing.

I agree with you except for this:

> This is not a setback for automatic driving. This is just a Tesla problem.

If the general public doesn't get the distinction it can become an automatic driving problem.

Fortunately we are at least 5 years away, probably more lie 10, from self-driving cars so this will be forgotten.

So for the next five years there won't be another fatal accident in which automatic driving is at least partially responsible and will generate significant press coverage?

You want to make a bet on that?

I think we're about 2-3 away from Google launching an Uber-like service in areas that lack snowfall.

You know that "5 years from mass production" is an euphemism for "never gonna happen," right?

Maybe in the short term, but in the long term, if automatic driving works, it works. There's just too much money to be saved by implementing it for people to not work on it.

As another example of this is Tesla "Bioweapons defense mode" which is pure hype and the testing they did in no way justified calling it that.

AFAIK, Tesla doesn't even have stereo cameras. It's a single, B&W camera behind the rear view mirror. The camera is used for object recognition, but the radar does all distance stuff.

It's rumored that the next version will have three cameras looking forward, so camera can be used for distance calculations.

I notice you comment on self-driving cars a lot and seem knowledgeable on the subject. Is this just something you follow closely or do you have some background in this area? I ask solely out of curiosity.