Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Any hack to remove the autopilot nag?

This site may earn commission on affiliate links.
In the long run I don't think torque sensing is the way to go, no matter how well it's tuned to the driver. There are scenarios where it just won't work well. Have it too sensitive, it'll be easy to trick and it will give false positives, on the other hand if it needs too much input, it'll just defeat the purpose of AP with the driver having to nudge the wheel all the time, even on long straight highway sections where AP should prove most helpful.

I also think a driver facing camera with a good recognition software is the way to go. That for sure will be able to tell if the driver is paying attention, it would be able to see the angle of the eyes, head position, blink rate, etc. It could tell if the driver is reclined too far or if from his position he can or cannot see the road. Really not hard to do, there are already eye trackers out in the gaming world and they can show exactly where the user is looking.

Maybe the camera together with some capacitive sensors in the wheel would be a good combo... but then people will start tying bacon strips to the wheel :p
 
I also think a driver facing camera with a good recognition software is the way to go. That for sure will be able to tell if the driver is paying attention, it would be able to see the angle of the eyes, head position, blink rate, etc. It could tell if the driver is reclined too far or if from his position he can or cannot see the road. Really not hard to do, there are already eye trackers out in the gaming world and they can show exactly where the user is looking.
Wow, that's like Orwell's 1984 in a Tesla.

IMaybe the camera together with some capacitive sensors in the wheel would be a good combo... but then people will start tying bacon strips to the wheel :p
Too funny...
 
  • Funny
Reactions: ElectricIAC
I'll take that as a "no" then; that you can't provide any information on these wonderful new attention confirmation methods.

I wasn't aware I was assigned to be your search monkey. But here you go...

https://www.eetimes.com/driver-monitoring-shut-up-mom-i-see-it/#

“VSI has been examining Tesla’s Autopilot for about six months, and we definitely see some weaknesses in Tesla’s approach, and this leads to system misuse. On the other hand, we recently got some seat time with GM’s Supercruise, and we see a substantially different approach. GM’s approach is more rigorous, which is consistent with their conservative implementation.”

"Earlier versions used steering-wheel sensors on the theory that a steering angle sensor could provide intelligence on driver fatigue. Tesla, on the other hand, uses torque sensors on steering wheels so that the vehicle knows whether a driver has hands on the wheel. “But for that, we have plenty of YouTube clips to prove how drivers can game the system,” said Barnden. Other systems use time-of-flight technology to infer a driver’s head position, he added."


So I'll challenge you to provide a link to any "peer reviewed" articles (or anyone with any credibility) stating that steering wheel torque sensors are a BETTER method of driver attention monitoring vs. cameras.


Tesla Considered Adding Eye Tracking and Steering-Wheel Sensors to Autopilot System

"One idea was sensors to track drivers’ eyes to ensure they watch the road. Tesla executives questioned the costs of such a system, which typically includes a camera and infrared sensor, and whether it would be ready for deployment, these people said."


Tesla Autopilot crash driver 'was playing video game'

"Tesla does instruct drivers to keep their hands on the wheel when using Autopilot, and an audible warning sounds if they fail to do so.

But the NTSB said "monitoring of driver-applied steering wheel torque is an ineffective surrogate measure of driver engagement".

I guess you disagree with the National Transport Safety Board?

Having one hand on the wheel to satisfy the sensor, while staring at your phone in your other hand. Would this guy still be alive if a camera was monitoring his attention?
 
Last edited:
I saw a video a while back by a guy who showed a proof-of-concept of another means to detect hands-on-wheel by, rather than relying on steady torque in one direction or the other, sending commands to the wheel to -- and this next word is important -- imperceptibly pull in alternating directions, and measure the resistance that way. Granted, it was a POC shown by the creator so of course it would appear to work well, but... it appeared to work well.
So your steering is going left but turning slightly right and vice versa. That would just confuse people and jam up the service centers.
 
some people who paid $7,000 dont think its a joke
they weren't expecting the 5+ year wait for return on their money....

Anybody who paid $7000 immediately got a return on their money- that's the post-March-2019 price, and it adds a slew of works right now features over basic AP (lane changing, Nav on AP, auto park, summon and smart summon, etc).

The only folks who've seen no return on their FSD purchase are the earlier buyers who only paid between 2-4k for it as an add-on to EAP (which already had all that stuff)



Agree, some folks really seem to lack a sense of humor.. take Knightshade, please, ba dum bum :D corrects and qualifies my joke... so sad.


Agreed- your joke was very sad indeed!
 
Last edited:
  • Funny
Reactions: ElectricIAC
.
As long as Tesla restricts autosteer speeds to 45-no-higher in places the database doesn't have speed limits, and 5-over-max anywhere it doesn't think is a proper spot for AP, then TACC will be an option for folks who are on a road where all traffic is going significantly faster than AP allows, especially where there might be some stop and go, to make driving significantly nicer/easier than not using anything.

Good point. Fortunately I don't get much of that terrain. The car's imposed "max allowed AS speed" is usually sensible. And as someone who considers AS/NOA the normal mode, like walking on hind legs is for humans, I encounter TACC as mostly that short dread moment when NOA ends, you're slowing off the exit, and just as you are taking over, if you don't execute the sequence perfectly, that &$#^# TACC demon takes off like the proverbial bat, blind and brainless, but very eager to hit higher speed, or Mr Lampost. Got to slap that puppy down. All in a day's drive. But TACC has a *** personality. I was going to use some adjectives, but in this day and age, somebody will always feel slighted. Anyway, not my fave.
.
 
@St☰v☰ Orwellian for sure. Your very last private place on earth gets policed, to be later swept up into the data banks (but WiFi only). :D

It looks like we have some very different views of society around here, where we draw the line. As we head for Knife Control and making sure you sit up straight, and don't look at the girls, that's not for me. Is it the fact that it's a computer that brings out this side of y'all?

You guys should mandate third party camera retrofits that easily install on the ignition line of ICE cars. Go after them first. They pollute and they misbehave a lot. If the camera detects "inappropriate behavior", it cuts the ignition circuit, and will not reconnect without a visit to the local police dept. Or the local mosque, depending on your registered address. The list of inappropriates is long, and expandable. In Zanzibaria monkeys are allegedly trained as Uber drivers to pull in some cash, illegitimately, and since they stopped printing phone books they sit too low; detect it!
.
 
.


Good point. Fortunately I don't get much of that terrain. The car's imposed "max allowed AS speed" is usually sensible. And as someone who considers AS/NOA the normal mode, like walking on hind legs is for humans, I encounter TACC as mostly that short dread moment when NOA ends, you're slowing off the exit, and just as you are taking over, if you don't execute the sequence perfectly, that &$#^# TACC demon takes off like the proverbial bat, blind and brainless, but very eager to hit higher speed, or Mr Lampost. Got to slap that puppy down. All in a day's drive. But TACC has a *** personality. I was going to use some adjectives, but in this day and age, somebody will always feel slighted. Anyway, not my fave.
.
I’ve found that TACC seems to not have so many “st3wpid” moments when disabling NoAP. But that one has to do this to remedy glaring problems with our expensive software is asinine. It really reinforces the decision to not “upgrade” on the Model Y.
 
  • Disagree
Reactions: anon125110
I wasn't aware I was assigned to be your search monkey. But here you go...

So I'll challenge you to provide a link to any "peer reviewed" articles (or anyone with any credibility) stating that steering wheel torque sensors are a BETTER method of driver attention monitoring vs. cameras.

...

But the NTSB said "monitoring of driver-applied steering wheel torque is an ineffective surrogate measure of driver engagement".

I guess you disagree with the National Transport Safety Board?

Having one hand on the wheel to satisfy the sensor, while staring at your phone in your other hand. Would this guy still be alive if a camera was monitoring his attention?

I dont have to provide anything or prove anything. It was you who made an unsubstantiated claim that "there were better ways" to monitor drivers, and its for you to provide supporting evidence. The VTI quote is interesting, but is just opinion. Has the guy done any actual research to quantify the different attention systems? I couldn't find anything to suggest this on his web site. So your claim remains, in fact, just your opinion.

As for the NTSB quote, its not relevant. The argument expressed by some of the posts in this thread more or less boils down to:

-- There are better systems than the one used by Tesla.
-- Therefore, the Tesla system is flawed.
-- Therefore, its ok to defeat it.

Which is nonsense, and has nothing to do with the NTSB opinion, nor the points I've been making about the dangers of nag defeat devices.
 
I'm one of them that paid full price in 2018 at delivery waiting for it to happen, don't get me started on the whole you'll get early access slow lol.


If you bought FSD in 2018 with delivery then you didn't pay anywhere near $7000 for it.

You paid $3000.

(also from everything I've seen you don't want early access. Hell half the time the initial public release of a new version breaks stuff. Better to set updates to regular (instead of advanced) and get the .2 or .3 version a week or two later with all the badly broken stuff patched)
 
If you bought FSD in 2018 with delivery then you didn't pay anywhere near $7000 for it.

You paid $3000.

(also from everything I've seen you don't want early access. Hell half the time the initial public release of a new version breaks stuff. Better to set updates to regular (instead of advanced) and get the .2 or .3 version a week or two later with all the badly broken stuff patched)
But they did pay for EAP+FSD as well as the generally higher price of the cars in 2018.
 
But they did pay for EAP+FSD as well as the generally higher price of the cars in 2018.
Exactly, we paid closer to $8000 if memory serves me right where you had to pay for enhanced autopilot separately for$5000 ( l if I remember right) then FSD was $3000 more on top of that, still best vehicle I ever had. I paid $3000 more than some people who got FSD and EAP for $5000, Elon said those of us that paid earlier would get earlier updates to compensate, never happened.
 
Last edited: