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

Tesla auto pilot 2 progress thread

This site may earn commission on affiliate links.
There’s been a lot of discussion and speculation on the progress of Tesla Autoplot 2.

As of August 2017, it still lags far behind Autopilot 1. However, it has been making steady, albeit. Frustratingly slow, progress.

So where do you think it will be by end of 2017? How about end of next year?

I want to link you guys to this: Patents by Assignee Mobileye Vision Technologies Ltd. - Justia Patents Search

It’s a list of MobilEye patents, many of which were granted this Summer. As an AI developer, I have wondered how much of the slow progress has been due to these patents and having to find a different, possibly more difficult, way to implement vision based systems. How hard, for instance, can it be to do pattern recognition for different types of vehicles to display them? I’d say the answer is it depends on the method used, and whether that method has been patented.

Here are my predictions:

I think they’ll achieve near AP1parity by end of this year. I don’t think we’ll get rain sensing, speed limit sign detection, vehicle type identification or showing cars in other lanes but I think general behavior will be pretty close. I think they’ll have to work out a licensing arrangement on patents to get the missing things and patent licensing can be slow.

I think we will need more specialized hardware (ap2.5) to get to real enhance d autopilot. Not because of raw computing power but because there will be a need to hardware accelerate patent-bypassing methods.

I think by this time next year we will have the enhanced auto pilot behavior we are hoping for. I don’t think we will see FSD for at least a year after that (2019) but it really depends on patents, not government regulation imo.

What do you think?
 
I hope you are right. I'm still a little salty on being given a 24 hour test drive with an AP1 car, asking if what I was purchasing was exactly the same as the test drive car, but it turns out I was not getting AP1 and instead getting an AP2 car.

I really loved seeing the parallel lanes and the related cars being show on the dashboard display. It helped put my mind to ease when engaging TACC and Autosteer and let me know the car was able to 'see' the other cars around me.

With AP2, I sometimes find the cone sensors not even picking up a car that's drifting into my lane until almost after I over take said car.

Other than that, AP2 has performed with above average performance peppered in with some pretty jarring and scary experiences:

1. Driving into the sun, TACC and Autosteer, engaged, it swerved hard to the right without any warning or anybody near me.

2. Random braking episodes when going under overpasses.

3. Issues with semitrucks, either getting too close or braking when passing them.

4. Issues with the map not being updated therefor the AP system slowing the car down to 45 on a freeway.
 
I think they’ll achieve near AP1parity by end of this year. I don’t think we’ll get rain sensing, speed limit sign detection, vehicle type identification or showing cars in other lanes but I think general behavior will be pretty close. I think they’ll have to work out a licensing arrangement on patents to get the missing things and patent licensing can be slow.
I think you put too much trust into Tesla's respect of IP of other parties.
If their licensing compliance is of any indication, they don't care at all about such minor details and would do anything, everything to keep their head above the water, consequences be damned (later!).
 
Hang on this doesn't make sense. Current EAP with AP2 hardware does not adjust speed on the fly. You must mean when first engaging TACC it gives you a slow start speed?

He means the TomTom map "geofencing" issue where on a freeway it may suddenly believe you're off-road or on a crossroad, resulting in a sudden speed limit restriction.

However, I hate to disappoint the OP, but many of the items pointed out are not uniquely AP2 regressions -- AP1 has had the same kind of issues especially as they've attempted to improve radar usage over time.
 
So much to learn. I had no idea it did anything like that as I've never seen it. It could also be region dependent since I'm in .au.
FWIW I’ve not personally experienced it on AP2 either. Similarly, judging from some of the posts here, I was afraid 2017.28 would try to kill me a few times a day. My experience has been far the opposite and from a road trip and commute point of view AP1 and AP2 have been quite similar.

It’s probavly worth remembering that not every issue affects every customer.
 
  • Informative
Reactions: GSP
Yes geofencing on invalid data is painful to say the least. There are 110 kph freeway sections north of Coffs Harbour that both TACC and AP2 will simply throw the brakes on without warning. It tends to happen when old roads that no longer exist intercept the new freeway. The car throws up messages "This road is limited to <speed> on autopilot".

Would love more current mapping data.

Car is tracking nicely along a freeway according to Google Maps, but the internal Nav is completely lost - generally thinks you're driving offroad.

Brendon
 
  • Informative
Reactions: conman
This is so painful to read :( Tesla wasted whole year trying to achieve the same functionality they had in 2015. The 2018 is nearing and once glorious and unique feature of Tesla is fading away being superseded by competitors.
To be honest I was considering buying Tesla a long time ago but was waiting for better AP, then TM3 cought my eye so I reserved one( should have same AP functionality as S) and was hoping for full autonomity ( How naive I was). Years have passed and not much has changed. AP 2 is still worse than AP1 and nothing from Tesla suggests it will get better soon( where is promissed coast to coast full autonomy drive?:( They are now focused on TM3 and Autopilot is not priority anymore. Eventhough I love Tesla I now find myslef look at other cars. I wanted Tesla mainly because of Autopilot but now when other car brands has the same functionality or even better I am not sure anymore.
I know I am just whining without even owning a Tesla but these are my thoughts as soneone who watch Tesla for couple of years now.
It is such a shame with the AP2, it looked so promissing;(
 
Lots of great points.

Honestly I’m hopeful that in the 9.0 release, parity will be complete and there will be EAP features as well. It took AP1 a year to fully function. This month would mark a year for AP2

9.0 should be out this month (October) based on previous .0 release times.
 
  • Funny
Reactions: pilotSteve
I wanted Tesla mainly because of Autopilot but now when other car brands has the same functionality or even better I am not sure anymore.(
I got my Model S on 9/28/17. I purchased Tesla mainly because of Autopilot. Buying electric was not a big deal to me since I work from home and do not put a lot of miles on my prior cars. I was really interested when I heard about the Cadillac CT6 Super cruise. And I read about the 2018 Audi and others. My daughter has a Honda Pilot with similar features (but not as good) so I assumed there must be other high end cars that are equal to or better then Tesla at even a lower cost. Too my surprise I could not find any that could match Tesla's Autopilot. So really not sure which cars you are talking about. To be clear I am talking about any car that I can actually BUY today (or last month) when I was making my decision.
 
I was at the dealer this weekend. My P85D lease will be up early next year. I am worried my new car next year will not be as good as my P85D - which has been perfect except the loose arm rest screw. AP1 works very well. I will cross my fingers AP2 will surpass AP1 by next year. If not I might just keep the P85D (I also like not having a center console).
 
  • Like
Reactions: ICUDoc
I hope you are right. I'm still a little salty on being given a 24 hour test drive with an AP1 car, asking if what I was purchasing was exactly the same as the test drive car, but it turns out I was not getting AP1 and instead getting an AP2 car.

I really loved seeing the parallel lanes and the related cars being show on the dashboard display. It helped put my mind to ease when engaging TACC and Autosteer and let me know the car was able to 'see' the other cars around me.

With AP2, I sometimes find the cone sensors not even picking up a car that's drifting into my lane until almost after I over take said car.

Other than that, AP2 has performed with above average performance peppered in with some pretty jarring and scary experiences:

1. Driving into the sun, TACC and Autosteer, engaged, it swerved hard to the right without any warning or anybody near me.

2. Random braking episodes when going under overpasses.

3. Issues with semitrucks, either getting too close or braking when passing them.

4. Issues with the map not being updated therefor the AP system slowing the car down to 45 on a freeway.

I think AP1 has the same issues, except perhaps random braking under overpasses.
 
This is so painful to read :( Tesla wasted whole year trying to achieve the same functionality they had in 2015. The 2018 is nearing and once glorious and unique feature of Tesla is fading away being superseded by competitors.
To be honest I was considering buying Tesla a long time ago but was waiting for better AP, then TM3 cought my eye so I reserved one( should have same AP functionality as S) and was hoping for full autonomity ( How naive I was). Years have passed and not much has changed. AP 2 is still worse than AP1 and nothing from Tesla suggests it will get better soon( where is promissed coast to coast full autonomy drive?:( They are now focused on TM3 and Autopilot is not priority anymore. Eventhough I love Tesla I now find myslef look at other cars. I wanted Tesla mainly because of Autopilot but now when other car brands has the same functionality or even better I am not sure anymore.
I know I am just whining without even owning a Tesla but these are my thoughts as soneone who watch Tesla for couple of years now.
It is such a shame with the AP2, it looked so promissing;(

Yes I basically feel the same way. There was so much potential all lost with the split from mobile Eye. The empty sockets on the AP2 circuitboard tell it all.
 
I drove my new MS with a friend who currently drives a Mercedes with autosteer functionality, and with friends who drove a Volvo with AS. Both were impressed with the smoothness of the Tesla autosteer. Mercedes as well as Volvo were much more unquiet, straight line as well as in bends, they said. And in the Mercedes you have to hassle the steering wheel every 15 seconds; Tesla requires 90 seconds, that's another big difference.
 
  • Informative
Reactions: conman
Comparing my first 4 days with AP2 to my year with AP1 I am actually quite pleased. Reading the various posts on here about what a disaster AP2 was caused some small bit of angst, and when initially saw the dancing lines in the IC I said "Oh no, here we go", but within 100 miles they settled down and the car has handled very well in autopilot. I took it to 7 distinct places that freaked out my P90D AP1 and it handled them all perfectly. Yes, I would like to see the cars in the other lanes, etc..., but I notice that despite not showing up in the IC, the car knows they are there when I try to auto lane change, and the lane change is both quicker and smoother than it was with AP1. The car seems to be smoother in braking in traffic, and smoother in keeping pace as the speed of traffic changes. To say that auto pilot in AP1 isn't keeping pace (heh heh heh) or that the move from MobilEye is a disaster seems to me to be completely wrong. Of course time will tell if AP2 continues to stay ahead and whether the added functionality that makes it enhanced will come into play. In any case, for me, I am already ahead of where my AP1 car was even in August.