Here's What's Missing from Self-Driving Cars: TRUST

[malcolm]

Not sure you're doing Tesla any favors by comparing them to Chipotle. You've obviously missed a few chapters in the Chipotle saga. My point simply was that companies develop a reputation for integrity, or food safety, or whatever. It doesn't mean they never make mistakes. But that reputation carries over into everything they do. GM certainly hasn't been a model company all the time. But at least the current leadership has tried to address serious issues in an honest and forthright way. Still waiting to see some of that from Elon & Co. And I'd want to see it before I took a nap in an FSD Tesla.

One man sees "honest and forthright" management while another sees appalling intransigence. Some people will never forgive GM no matter what they do. They will always be the EV1-crushers or the ignition-switch-killers.

Takata are the air-bag-killers.

For some, their personal experiences with Tesla's Autopilot have left them shaken and there is simply no way back from that.

Some set particular targets - for you, it seems to be some non-specific act of contrition from Elon, which GM's management have (apparently) already met.

Although, maybe not:

Would I TRUST Tesla to make a safe self-driving AP2 car without me in total control? Probably never.

And so we praise the next generation of Autopilot from Audi, BMW, Mobileye - whoever - simply because any future product is always untarnished by the both the reality of fallible, all-too-human manufacturing and the real-world experiences of customers.

Yes it's good to dream of a idealised future, but the idealised future is always the truest form of vaporware.

 

[NerdUno]

Time will tell, @malcolm. Doesn't look like we'll have to wait very long. There's a reason a dozen different companies are working on this, and it's not to produce vaporware.

 

[MarcusMaximus]

I haven't bothered to read any replies from my last post, because of the fanboi brain turds. But today I reached a new level of infuriation. Looks like the latest firmware 17.24.28 lies even harder than it used to. An easy ~35 mile round trip in cloudy ~75F weather, the trip meter usage was off by a whopping 12%! Looks like they're trying even harder to report good consumption even though real consumption was terrible. Hey, if you're the kind of owner that doesn't use loggers and doesn't check their consumption, you just get home and plug in, thinking you had an efficient trip, because the trip meter told you so right?

But it's LIES LIES LIES.

Software is supposed to get better over time, apparently it only improves to make Tesla look good at the cost of all honesty. Why not just tell me my consumption was 200Wh/mi? Look how efficient my Tesla is...

BLAH

What is a "trip meter usage"? It's hard from this post to tell what you're talking about or why you're upset.

 

[stopcrazypp]

Are you kidding me right now? Tesla AP 2 does the same thing. EVEN WORSE.

There are dozens of fail videos where AP has you comfortably in the other lane repeatedly.
Seriously you tesla fans are killing me with your buffoonery.


Volvo Pilot Assist 2 is FAR better than AP2 and equals and surpasses AP1 in many avenues.

Here comes your excuses and patented deflections in 3...2...1

Par for the course for you. Posting videos of cars in different conditions is meaningless. Those aren't objective reviews done under the same conditions. And one system better than AP2 or AP1 does not change the point. My point was that there are plenty of "non-beta" systems that do worse than Tesla's under the same conditions. Having the "beta" label or not is not a indicator of quality.

As for PA2, I looked at the forum reviews. It seems a ton worse than AP. It can keep the car in the lane when the road is almost straight, but struggles for even highway curves (much less undivided windy local roads with intersections).
Also, hands on wheel detection is even worse than Tesla's it seems.
Pilot Assist 2 - working as designed or faulty?
Auto Pilot Assist 2 thread (experience, problems) - Page 5

Waiting for an objective comparison between the two (and also Supercruise). The car magazines will probably do one if PA2 really is the cutting edge.

 

[AWDtsla]

What is a "trip meter usage"? It's hard from this post to tell what you're talking about or why you're upset.

Trip meter says consumption rate X, consumption Y. Real consumption rate is 1.12*X, real consumption is 1.12*Y. Check all the other threads about pack capacity and calculating usage from rated miles, etc. There's plenty of details and plenty of posts there.

Imagine if BMW said you were going to get 30 MPG, you get in the car and it often says 30 MPG. But then you decide to measure out the gasoline, and it turns out it only gets 26 MPG when it reports 30. Class action time.

 

[AWDtsla]

Tesla is not getting the message. There digging their grave deeper.

 

[stopcrazypp]

Trip meter says consumption rate X, consumption Y. Real consumption rate is 1.12*X, real consumption is 1.12*Y. Check all the other threads about pack capacity and calculating usage from rated miles, etc. There's plenty of details and plenty of posts there.

Imagine if BMW said you were going to get 30 MPG, you get in the car and it often says 30 MPG. But then you decide to measure out the gasoline, and it turns out it only gets 26 MPG when it reports 30. Class action time.

I guess you didn't know that fuel economy gauges are inaccurate in general and they all err on the side of reporting high (19% higher in one case)?
Your Fuel Economy Gauge Is Fibbing

Good luck with that class action.

 

[Shorty]

^ Well if the gauges reported low then someone would want to sue over that claiming that they had higher evaporation due to having to buy more fuel! You can sue anyone for anything, you don't even have to have a logical argument. Got to love how the system works.

Of course what the lawyers often 'forget' to tell their clients is that in some cases the defendant presses for costs when they are successful and the law firm isn't covering those!

 

[b team]

Trip meter says consumption rate X, consumption Y. Real consumption rate is 1.12*X, real consumption is 1.12*Y. Check all the other threads about pack capacity and calculating usage from rated miles, etc. There's plenty of details and plenty of posts there.

Imagine if BMW said you were going to get 30 MPG, you get in the car and it often says 30 MPG. But then you decide to measure out the gasoline, and it turns out it only gets 26 MPG when it reports 30. Class action time.

False analogy.

 

[verygreen]

Have you looked into the differences between 17.24.30 and 17.24.28 firmware files?

So the differences are minimal as expected. So minimal in fact that even the image file sizes are the same between the two.
Every binary is still slightly different because they don't have repeatable builds at Tesla, apparently, and embed timestamp/id into every binary as it gets built.
17.24.30 was built on June 27th evening.

 

[alehbaba]

no way i can trust any computer to drive my car around....never.....who will you blame when you get smashed up? someone gets killed?...car gets dinged? the computer? tesla? lol.....i can see why insurance will go skyhigh with ap cars...

 

[Bladerskb]

Par for the course for you. Posting videos of cars in different conditions is meaningless. Those aren't objective reviews done under the same conditions. And one system better than AP2 or AP1 does not change the point. My point was that there are plenty of "non-beta" systems that do worse than Tesla's under the same conditions. Having the "beta" label or not is not a indicator of quality.

As for PA2, I looked at the forum reviews. It seems a ton worse than AP. It can keep the car in the lane when the road is almost straight, but struggles for even highway curves (much less undivided windy local roads with intersections).
Also, hands on wheel detection is even worse than Tesla's it seems.
Pilot Assist 2 - working as designed or faulty?
Auto Pilot Assist 2 thread (experience, problems) - Page 5

Waiting for an objective comparison between the two (and also Supercruise). The car magazines will probably do one if PA2 really is the cutting edge.

We already went through this before the person who started that same thread started praising PA2 in the end once they figured out a hack that lets him keep PA2 engaged forever. Infact he talked about going hundreds of miles without touching the wheel.

The problems being mentioned there all has to do with the small timeout (about 15 secs).
while in comparison there are tens of thousands of anti-AP posts listing its problems, from truck lust to ramming cars infront, to phantom deadly brakes over overhead signs and such.

works only on complete straight road?

lol what is this video right here faked?

or this one? so many fake vids.

I posted you video of PA2 handling undivided local roads.

same road type with AP2 in comparison. the poor driver barely made it alive.

sorry that i don't have 9,000 vid like tesla cults have posted to demo PA2 supremacy but try harder bro.

Lastly i knew you would come back with deflection. this is not a different condition. you claimed that AP won't put you in another lane. Well it clearly does. game set match.

PA2 surpasses AP1 although barely and Supercruise easily triumphs AP1 by far (its not even close). Both of them though destroy AP2.
Its not even debatable.

 

[malcolm]

game set match.

You forgot to mention how cheap it is to upgrade.

Owners discuss Volvo's Pilot Assist II upgrade

And, of course, Super cruise isn't out yet.

 

[AnxietyRanger]

Par for the course for you. Posting videos of cars in different conditions is meaningless. Those aren't objective reviews done under the same conditions. And one system better than AP2 or AP1 does not change the point. [...]

As for PA2, I looked at the forum reviews. It seems a ton worse than AP. It can keep the car in the lane when the road is almost straight, but struggles for even highway curves (much less undivided windy local roads with intersections).

I think this is the part where @stopcrazypp the fact that you don't own an AP2 car is starting to really hurt the argument - or at least you should be really careful when implying stuff that might include AP2. I know you just say AP, but given that most here actually praise AP1 and AP2 is where the really big criticism is at, it sounds inaccurate.

AP2 is terrible in undivided roads - and a disaster in intersections you mention - and it, really, is not that good at driving divided roads either... and there you make it sound completely different with a quote like that. I have no idea how PA2 drives, but I do know how AP2 drives and your description does not sound accurate.

You sounding like simply taking the good of AP2 and the bad of PA2 just looks like you have a very strong bias or agenda. I'd advise more care.

p.s. I have no problem with the beta label or that discussion. IMO it is sort of a red herring.

 

[McRat]

...
And, of course, Super cruise isn't out yet.

It should be here in Sept. Finally. After 5+ years. It was demonstrated in early 2012 to the press. Cadillac testing 'Super Cruise' feature for future cars. So it was probably started much earlier. This was pre-Model S. Sure, Cadillac could have released it and updated it with OTA updates (OnStar based OTA has been happening for about a decade to little fanfare), but I'd guess that GM is risk-adverse when it comes to litigation.

Read and watch:

Putting Cadillac's self-driving Super Cruise technology to the test

Much of the recent delay has been regulatory. But after the NHTSA decided running into trucks is an acceptable behavior for a self-driving feature, Cadillac can now proceed.

That being said, my wife did not want that feature in a car. The ability of a car to encourage distracted driving is not a safety feature, it's a safety hazard. AEB/ACC/LKA/BSA are fine. Pretending the car will always make the correct life-or-death decision better than a experienced driver has not been supported by any company yet.

 

[Bladerskb]

Much of the recent delay has been regulatory. But after the NHTSA decided running into trucks is an acceptable behavior for a self-driving feature, Cadillac can now proceed.

This is 100% incorrect. I don't know where people get this narrative from. its not true. the regulatory inquiry from NHTSA was simply about GM saying they will turn hazard lights on when the car stops and that was after GM notified them about their system summer last year and got a response Nov/Dec. None of the delays have anything to do with regulatory.

 

[McRat]

This is 100% incorrect. I don't know where people get this narrative from. its not true. the regulatory inquiry from NHTSA was simply about GM saying they will turn hazard lights on when the car stops and that was after GM notified them about their system summer last year and got a response Nov/Dec. None of the delays have anything to do with regulatory.

I think you are referring to the informal rejection of the automatic emergency flasher operation per FVMSS regulations, then submitted formally as a rebuttal by GM on March 18, 2016. The informal ruling was later changed by the NHTSA to a conditional formal acceptance if GM can insure the system is safe dated April 20, 2016 by Paul Hemmersbaugh, Chief Council of the NHTSA.

This does not mean no informal communications occurred prior to the official submission on March 18, 2016 for a formal ruling. In fact, there must have been. Those docs I cannot find without a FOIA request. The dates you list must be informal communications.

The question remains that if the NHTSA gave a conditional, why the 2016 (late release) Cadillac nor the 2017 Cadillac were fitted with the then operational system. In all likelihood it was more informal discussions to remove the conditional which opens GM up to risk.

"Can I build this car?" "Sure as long as you guarantee it's safe." "Uh... Cars aren't safe." "Sucks to be you!."

 

[Bladerskb]

I think you are referring to the informal rejection of the automatic emergency flasher operation per FVMSS regulations, then submitted formally as a rebuttal by GM on March 18, 2016. The informal ruling was later changed by the NHTSA to a conditional formal acceptance if GM can insure the system is safe dated April 20, 2016 by Paul Hemmersbaugh, Chief Council of the NHTSA.

This does not mean no informal communications occurred prior to the official submission on March 18, 2016 for a formal ruling. In fact, there must have been. Those docs I cannot find with a FOIA request. The dates you list must be informal communications.

The question remains that if the NHTSA gave a conditional, why the 2016 (late release) Cadillac nor the 2017 Cadillac were fitted with the then operational system. In all likelihood it was more informal discussions to remove the conditional which opens GM up to risk.

"Can I build this car?" "Sure as long as you guarantee it's safe." "Uh... Cars aren't safe." "Sucks to be you!."

This is 100% false. First of all, everytime that GM delay supercruise they cited technical problems. They talked sun glare, etc.

second of all, ALL official NHTSA letters are available online, nothing is hidden. it has nothing to do with whether the car is safe or isn't safe. but everything to do with the hazard lights.

Here is the letter: 16-1289 (GM hazard innovative) -- 28 Apr 16 rsy

This responds to your letter dated March 18, 2016 requesting an interpretation with respect to the meaning of vehicle hazard warning signal operating unit in Federal Motor Vehicle Safety Standard (FMVSS) No. 108; Lamps, reflective devices, and associated equipment, as applied to a new cruise control system General Motors (GM) is developing.

You state that GM is developing a new adaptive cruise control system with lane following (which GM has referred to as Super Cruise) that controls steering, braking, and acceleration in certain freeway environments. When Super Cruise is in use, the driver must always remain attentive to the road, supervise Super Cruise's performance, and be ready to steer and brake at all times. In some situations, Super Cruise will alert the driver to resume steering for example, when the system detects a limit or fault. If the driver is unable or unwilling to take control of the wheel (if, for example, the driver is incapacitated or unresponsive), Super Cruise may determine that the safest thing to do is to bring the vehicle slowly to a stop in or near the roadway, and the vehicle's brakes will hold the vehicle until overridden by the driver.

You indicate that GM plans to develop Super Cruise so that, in this situation, once Super Cruise has brought the vehicle to a stop, the vehicle's automated system will activate the vehicle's hazard lights. You state that you believe that this automatic activation of the hazard lights complies with the requirements of FMVSS No. 108 for several reasons.

We agree with GM that the situation it describes is similar to the situation in which the Steele (and Bartlett) letters that interpreted FMVSS No. 108 to permit automatic actuation of the hazard lights..Therefore, the automatic activation of the hazard lights in the circumstances described by GM would be permissible.

We urge GM to fully consider the likely operation of the system it is contemplating and ensure that this fallback solution does not present an unreasonable risk to safety

This was in response to the first letter GM sent to NHTSA telling them about their system and how it operates and inquiring about a specific law.
There is nothing to see here, there is no regulatory issue, block, hold up, delay. Can ppl stop coming up with fairy-tales for once, for once. my goodness.

This is a non-issue, its exactly what autopilot and other driver assistance have been doing for years, they turn on the hazard lights.

lastly super-cruise was delayed before GM ever even talked to NHTSA about their new system (jan 2016).

GM delays Super Cruise rollout
game set match.

 

[McRat]

...
game set match.

You should not edit rulings. The letter you quoted from the NHTSA to GM includes this passage:

Past agency interpretations of automatic activation of hazard lights have reached different conclusions about their permissibility. Some past agency interpretations of FMVSS No. 108s definition of hazard warning signal operating unit have construed it to preclude automatic operation of the hazard warning lights, on the basis that automatic activation would not be driver controlled.

Believe whatever blogs you wish, but do not edit official docs to alter them to your viewpoint. Bad dog.

What do you think a Past Agency Interpretation means? That it will be found in the next paragraphs of the document? Find the Past Agency Interpretation. Good luck.

 

[Bladerskb]

You should not edit rulings. The letter you quoted from the NHTSA to GM includes this passage:

Past agency interpretations of automatic activation of hazard lights have reached different conclusions about their permissibility. Some past agency interpretations of FMVSS No. 108s definition of hazard warning signal operating unit have construed it to preclude automatic operation of the hazard warning lights, on the basis that automatic activation would not be driver controlled.

Believe whatever blogs you wish, but do not edit official docs to alter them to your viewpoint. Bad dog.

What do you think a Past Agency Interpretation means? That it will be found in the next paragraphs of the document? Find the Past Agency Interpretation. Good luck.

Whose edit rulings? me or you? i posted what was relevant.
and in response to past agency interpretation.

"However, since those interpretations were issued, NHTSA has clarified that automatic activation is permissible in certain circumstances. In a 2002 interpretation letter issued to Bartlett Industries, Inc., NHTSA explained that the hazard lights may be automatically actuated following a vehicle crash:"

These past interpretations are OLD and already superseded by NHTSA clarification in 2012.

aka STOP SELECTIVE READING

In conclusion, gm supercruise was delayed december 2015 to Fall 2017 for technical reasons before GM even contacted NHTSA to tell them about their system in march 2016 and specifically inquiring about hazard lights which NHTSA agreed with them about. At no time were there regulatory blocking, hold up, delay or any such nonsense that gets spilled around here constantly.