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

Navigate on Autopilot is Useless (2018.42.3)

This site may earn commission on affiliate links.
13657DD4-B4D5-4BD7-A57E-483E88B904CA.jpeg


Nice to see the meme lives on in Tesla. Bonus points to whoever wants to photoshop it to have Tesla’s logo.
 
  • Funny
Reactions: boonedocks
Starting in 2016 - our 1st 2 teslas were AP II. Then we got a loaner that was AP1 - that 1st year (air suspension time out - module failed before we even got our S home). That old fashioned mobileye acted /felt like we'd just traded up for something more advanced. There's no point in telling people their newer cars are primitive in comparison - because they often just don't get it - unless they've driven thousands of miles in both. But no one can overcome cultlike defensive thinking. So yea - of COURSE folks will vehemently disagree. Maybe many here don't realize it, but some of us got a pittance of a class action settlement because the difference was demonstrable - & tesla settled out so it wouldn't become a giant brouhaha. Being a fan, I'm glad they did.
Still - it's a pity. One can only imagine how far Tesla's full self-driving would be had the divorce between the two companies not happened.
.
 
Last edited:
  • Disagree
Reactions: Electroman
Gents - keeping the hand on the wheel does not matter. What is needed watch the road, and be alert. That is enough to take over before any problem happens. Keeping the hand on the wheel alone does not help.

Alertness is by far the most important aspect, but having hands and feet at or immediately over the controls is also critical. Human reaction times are about 250ms, and to move your hand or foot several inches after initial recognition of an emergency is around a second. On top of that, you have panic setting in which often causes pressing the wrong pedal.

Tesla is measuring attentiveness wrong, but it's not right to say that an attentive drive with hands off controls is a good idea either.
 
It can very easily sense if it you let gravity apply pressure on the wheel. Do that and you never need to apply force and the car never nags. I do this every single day.
My regular touch is too supple for consistent detection. I often glance down and see the hands on warning box when my hands have been on the wheel for many miles. Putting the steering in sport mode helped a bit. I still have the car hesitate with automatic lane changes until I give the wheel a bit of a shift.
 
  • Like
Reactions: rnortman
Tesla is measuring attentiveness wrong, but it's not right to say that an attentive drive with hands off controls is a good idea either.

Agreed that hands on wheel is a good idea. I'm not sure what a good/better way to measure attentiveness is though. I'm dubious of any AI camera based system. Sure, it can figure out when you are asleep, or staring out the side window for long periods. But what about when you just space out staring ahead listening to music?

Various suggestions have been made, including just recently a politician chiding Tesla for not doing more. But at the end of the day, people will always find a way to game the system (weights on the steering wheel, anyone?). As my father used to say, never underestimate the ingenuity of stupid people.
 
  • Like
Reactions: DrDabbles
Starting in 2016 - our 1st 2 teslas were AP II. Then we got a loaner that was AP1 - that 1st year (air suspension time out - module failed before we even got our S home). That old fashioned mobileye acted /felt like we'd just traded up for something more advanced. There's no point in telling people their newer cars are primitive in comparison - because they often just don't get it - unless they've driven thousands of miles in both. But no one can overcome cultlike defensive thinking. So yea - of COURSE folks will vehemently disagree. Maybe many here don't realize it, but some of us got a pittance of a class action settlement because the difference was demonstrable - & tesla settled out so it wouldn't become a giant brouhaha. Being a fan, I'm glad they did.
Still - it's a pity. One can only imagine how far Tesla's full self-driving would be had the divorce between the two companies not happened.
.
I have had both. The early iterations of AP2 was definitely inferior to AP1. But every update it got better, and I can confidently say the current version of AP2 is much superior to AP1 in every aspect.

AP1 as good as it was, it was a dead end. AP2 with all the initial struggles, now with 7 cameras is not only better but has got so much potential to be unlocked.
 
I'll be making no such thing. lol I'd just hate to see someone get their hands completely mashed by an exploding airbag. They're much more powerful than people expect, and if your hand or arm is tin the way, you can be severely injured. Even the powder they used to be packed with was causing injury to people, to give a sense of how strong a deployment is.
 
I'll be making no such thing. lol I'd just hate to see someone get their hands completely mashed by an exploding airbag. They're much more powerful than people expect, and if your hand or arm is tin the way, you can be severely injured. Even the powder they used to be packed with was causing injury to people, to give a sense of how strong a deployment is.
The world might not deserve a Dr. Dabbles youtube channel, but it certainly needs one.
 
I'll be making no such thing. lol I'd just hate to see someone get their hands completely mashed by an exploding airbag. They're much more powerful than people expect, and if your hand or arm is tin the way, you can be severely injured. Even the powder they used to be packed with was causing injury to people, to give a sense of how strong a deployment is.

I can attest to that from personal experience, when some years back I was in a car that had an airbag deployment (long story). The airbag blew the (detachable) sunroof off the car, took my glasses off my face and sent them flying out the sunroof opening, and ripped my wristwatch (steel band) off my hand and took that out the sunroof too. OTOH I stepped out the car after a 50mph head-on collision (the guy drove into me) with no whiplash etc and only a scar on my wrist where the wristwatch had been.
 
  • Like
Reactions: DrDabbles
Starting in 2016 - our 1st 2 teslas were AP II. Then we got a loaner that was AP1 - that 1st year (air suspension time out - module failed before we even got our S home). That old fashioned mobileye acted /felt like we'd just traded up for something more advanced. There's no point in telling people their newer cars are primitive in comparison - because they often just don't get it - unless they've driven thousands of miles in both. But no one can overcome cultlike defensive thinking. So yea - of COURSE folks will vehemently disagree. Maybe many here don't realize it, but some of us got a pittance of a class action settlement because the difference was demonstrable - & tesla settled out so it wouldn't become a giant brouhaha. Being a fan, I'm glad they did.
Still - it's a pity. One can only imagine how far Tesla's full self-driving would be had the divorce between the two companies not happened.
.


I agree to some facts on AP1, however I drove a loaner on somewhat of a hilly road to my house and it nearly wrecked the car by wanting to cross left into the oncoming lane on autopilot, my AP2.5 and 3.0 navigate these hills daily without any excitement.
 
I don't find i1Tesla testing on a narrow two lane country road and no shoulder road useful for Naviage on Autopilot (NoA) testing to be valid [subject of thread]. NoA is for highway ramp to ramp like the video I posted is doing.
ie. NoA = 1 blue line vs normal AutoPilot = two blue lane lines.
Hmmmm, I have the same experience as in the video above except that I am driving on a road MUCH worse than what he is using. Sometimes with no right line and often a faded center line. As of that same update, my car drives the road perfectly where as I would have never used it on that road in the past. It would also slow down on tight curves as needed. I guess it all depends on the roads. Yea, I know it is mainly for highway driving.