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

Just a Place Keeper to See if I Am Right About the Future of AP1

This site may earn commission on affiliate links.
I agree entirely. As I said in a post a couple of weeks ago: I know I may be a heretic, but I am not surprised that restrictions on AP 1.0 are coming. It was always beta software and its limitations will become increasingly pronounced as AP 2.0 matures -- the future is that AP 1.0 will become non-supported legacy hardware/software. I would not be surprised that it will be forced into retirement by regulators -- this would probably be good for the safety of the public and for Tesla. I won't miss it much. The best part of owning a Tesla, in my view, is driving one.
But the point is that we didn't pay $2500 for a system in beta. The advertisement didn't read "for $2500, you agree to take part in our autopilot beta program. It might work out, it might not. But you can be the first to try it out." I would think that, had the feature been promoted this way, there would still be a large percentage of us who would have ponied up just to try it out. But that's not what was being sold. We paid for a fully operational autopilot. If this is never delivered, Tesla will have to make amends.
 
When Tesla comes to an agreement "with" the NHTSA to make changes to AP1.0 the customer (you) have no say in the matter.
So we've re-defined "Mandatory" along with "recall", "order", and several other words.

Just so you know, the NHTSA has never, not a single time in the entire history of automobiles, made a recall mandatory for the consumer. They make the manufacturers offer, but the end customer is allowed to refuse. You are stating that this will be the first time. And not only that, but that the NHTSA won't even tell us about it.

Anyway, I'm done. Continue to blabber on about your conspiracy theory. keep ranting about non-mandatory non-order non-recalls that came out of super secret meetings. I will now give them the attention they deserve. Right up with flat earths, 9-11 inside jobs, and faked moon landings.

I own my car. There is nothing Tesla can legally do to remove ANY functionality they previously delivered to me under any circumstances. EVEN if the the NHTSA did order it (because they have no jurisdiction in Canada, and the law says even the manufacturer can't make ANY changes to my car without my authorization) And I don't have to worry about Transport Canada ordering it, because they don't actually have the authority to make recalls mandatory at all!

Feel free to tell me to sue Tesla then, it may yet come to that, but I'm following the proper processes and leaving the lawsuit as a last resort rather than a first recourse. I have gone as far as Tesla will go (which isn't very far by the way, I couldn't even talk to anyone with the authority to discuss the issue, let alone resolve it) and am now starting with regulatory agencies. I'll know more in a month or two.
 
The formal paperwork signed before any negotiations make it very clear that all negotiations are without prejudice and confidential. Only at the conclusion can disclosure take place. Disclosing agreements made during the course of the negations risks the breakdown of the entire negotiations.

So...we agree then?

But the point is that we didn't pay $2500 for a system in beta. The advertisement didn't read "for $2500, you agree to take part in our autopilot beta program. It might work out, it might not. But you can be the first to try it out." I would think that, had the feature been promoted this way, there would still be a large percentage of us who would have ponied up just to try it out. But that's not what was being sold. We paid for a fully operational autopilot. If this is never delivered, Tesla will have to make amends.

I'm pretty sure we did. I seem to recall this being all over the initial announcement as well. When I took Delivery in May 2016, my Delivery Specialist was very clear that the system was in beta and cautioned me to read the entire agreement before enabling the system. Change is the nature of a product in Beta. I suspect we will see more changes as both AP1 and AP2 mature.
 
On what do you base this assertion? I keep hearing over and over again that it "isn't safe to go slower than the flow of traffic." Is it? Or is this just the perception of those who want to rationalize speeding? Please show me the studies on which this assertion is based.

Do you like, drive a car? Are you the guy doing the speed limit in the passing lane?
 
You can also go only a few seconds under some situations, like accelerating to pass.
Not to mention, once the nags are that frequent, it almost doesn't matter exactly how frequent, the result is you have to have a weight hanging from one side of the steering wheel at all times to satisfy it.

There's a huge difference between a system that nags every minute under all conditions and one that nags that often in corner cases.

The advantage to the MB system is that it works on all roads at any speed, and doesn't penalize you if you happen to miss a nag 3 times in an hour.
Again, unless the MB system has improved vastly in the past few months, it doesn't matter at all that it can be engaged in all circumstances. It just doesn't work very well. I get your philosophical argument but, personally, I'd much rather have a system that can be activated in limited circumstances (say, divided highways) and works well under those situations than one that can be activated under all circumstances but doesn't work well in any of them.
 
Last edited:
  • Like
Reactions: NOLA_Mike
There's a huge difference between a system that nags every minute under all conditions and one that nags that often in corner cases.


Again, unless the MB system has improved vastly in the past few months, it doesn't matter at all that it can be engaged in all circumstances. It just doesn't work very well. I get your philosophical argument but, personally, I'd much rather have a system that can be activated in limited circumstances (say, divided highways) and works well under those situations than one that can be activated under all circumstances but doesn't work well in any of them.

I assume you have first hand experience driving the E300 with Drive Pilot?

I can only speak on what I have experience with, and regarding Tesla's AP I have practically none aside from using it about 3 minutes on my test drive. I do however have experience with Mercedes' system and if Tesla's would behave exactly the same, I would be happy.

I find that the system works well on straight well marked roads, both highway and surface streets and handles gentle curves well. TACC does a good job of keeping the proper distance and comes to a full stop and starts back up again when needed. The system also changes lanes by itself once initiated with the turn signal stalk, similar to Tesla's.

The system does struggle with aggressive curves and needs the drive to take over and it does nag you every 30 seconds to grab the wheel. I do not see that as a problem though and quite normal actually, but then again I've always viewed these systems as driver assistance features where others seem to think of them as self driving (with the driver sitting as a passenger). That really isn't something I personally need it to do. I would just like some assistance, maybe to teach and get something or read a quick text.

Also, one thing I think is different between Drive Pilot and Autopilot is that DP stays active even if you take over to manually street temporarily while I believe AP turns off in that type of situation.
 
I'm pretty sure we did. I seem to recall this being all over the initial announcement as well. When I took Delivery in May 2016, my Delivery Specialist was very clear that the system was in beta and cautioned me to read the entire agreement before enabling the system. Change is the nature of a product in Beta. I suspect we will see more changes as both AP1 and AP2 mature.
No, it does not say beta on my delivery invoice. I accept the fact that the features would be rolled out over the upcoming months (starting in October of 2015). I accept the fact that there would be a beta period. However, this cannot go on indefinitely. Eventually, Tesla has to offer the advertised features.

To be clear, I am a big proponent of AP1. It does as well if not better on the highway than I ever expected. There was never any promise of it working on secondary roads and city streets, and yet it does pretty well. I have no complaint about its ability here as this was a feature that was never promised.

However, summon was promised. Summon is still in beta. Summon does not do what was advertised. Summon needs to improve. I am very skeptical it ever will. If Tesla can't deliver, what recourse should we have?
 
  • Like
Reactions: robby and u00mem9
Who says the Mercedes auto pilot doesn't work well? It will auto steer even without lane markings up to 100 mph. That is on the new E class.

And also it swerves around obstacles as my understanding, not slam on the brake as someone had mentioned.

The 2017 Mercedes-Benz E300 challenges Tesla's Autopilot with Drive Pilot | TechRadar

I can attest that it does not work well around even moderate curves. I tested the system out on this stretch of road going home one day and it could not navigate it. I was in the middle lane (3 lanes each way, with a center concrete divider) and the car barreled into the next lane and was headed straight for the outside concrete barrier before I took over control.

Screen_Shot_2017_01_03_at_8_49_56_PM.png


Now I know that it cannot handle that situation and know to take full control when I see a curve coming up. Luckily most of the Phoenix area is very straight north, south, east and west roads and in a straight line, the system does well to keep the car centered.
 
  • Informative
Reactions: somnambule
I assume you have first hand experience driving the E300 with Drive Pilot?

I can only speak on what I have experience with, and regarding Tesla's AP I have practically none aside from using it about 3 minutes on my test drive. I do however have experience with Mercedes' system and if Tesla's would behave exactly the same, I would be happy.

I find that the system works well on straight well marked roads, both highway and surface streets and handles gentle curves well. TACC does a good job of keeping the proper distance and comes to a full stop and starts back up again when needed. The system also changes lanes by itself once initiated with the turn signal stalk, similar to Tesla's.

The system does struggle with aggressive curves and needs the drive to take over and it does nag you every 30 seconds to grab the wheel. I do not see that as a problem though and quite normal actually, but then again I've always viewed these systems as driver assistance features where others seem to think of them as self driving (with the driver sitting as a passenger). That really isn't something I personally need it to do. I would just like some assistance, maybe to teach and get something or read a quick text.

Also, one thing I think is different between Drive Pilot and Autopilot is that DP stays active even if you take over to manually street temporarily while I believe AP turns off in that type of situation.

My experience with Drive Pilot was brief and a few months ago (which is why I kept adding the unless-it's-improved-in-the-past-couple-of-months caveat) but what you described is consistent with my perception.

Tesla's AutoPilot, in my opinion, is far more robust at staying within a lane including through curves. Not only does it stay within the lane, but stays centered almost all the time, which inspires a lot more confidence, at least for me. The nags, as I noted, are also less frequent in the Tesla system. I agree that you are always required to hold the wheel with either system but the reason I pointed that out is that the increased frequency of nags has been a major source of complaints here.

As you noted, one of the biggest design differences is that DP doesn't disengage when you manually steer. I think that's consistent with MB's positioning of DP as a potentially "always-on" driving aid and Tesla's may be more consistent with where they want to go with AP with the car taking over progressively more responsibilities but still providing an easy and intuitive "override and cancel". I do wish AP had a mode where manual intervention didn't always cancel it, but given how easy it is to re-engage, it's not that big a deal.
 
  • Like
Reactions: kevinf311
From reviews that Ive read on both systems, even compared head to head in a review, the Tesla system came up way on top - but again as was mentioned by someone else, this was when Tesla's system was far more liberal in it's use and a lot less restrictive. And at this point, Tesla's is not anywhere near as robust as Mercedes...as it doesn't work AT ALL. :p

Not that it matters much, but I would love to have the restricted version of what AP1 cars have right now...hell even nagging every 30 seconds would be fine...hell even a version of Mercedes would be fine. I would love it as a drivers assistance, it does not need to drive me around and I am not looking to read the newspaper while the car drives me down the road.
 
From reviews that Ive read on both systems, even compared head to head in a review, the Tesla system came up way on top - but again as was mentioned by someone else, this was when Tesla's system was far more liberal in it's use and a lot less restrictive. And at this point, Tesla's is not anywhere near as robust as Mercedes...as it doesn't work AT ALL. :p

Not that it matters much, but I would love to have the restricted version of what AP1 cars have right now...hell even nagging every 30 seconds would be fine...hell even a version of Mercedes would be fine. I would love it as a drivers assistance, it does not need to drive me around and I am not looking to read the newspaper while the car drives me down the road.

Any day now, right? I would expect your EAP/FSD car would be passing our AP1's capabilities within a couple of months.
 
HX_guy, I trust your opinion. Here's an unbiased YouTube video that confirms what you are saying. It is great at keeping distance and braking, good on straight roads, a bit questionable on some curves. I wonder if it needs adjacent cars to do a better job.


Here are the claimed features on the 2017 E-class:


  • Drive Pilot’s adaptive cruise control (Distronic) and Steering Pilot (lane keep assist) track the car in front at speeds up to 130 mph (210 kph) and can stay lane-centered at up to 81 mph (130 kph). Mercedes says the car can track adjacent vehicles and deal with obscured (on one side) lane markings. In general, a car that can do this at incredibly high speeds provides a greater margin of safety on US roads. Where most steering-assist cars demand the driver put hands back on the wheel every 10-15 seconds, the E-class will go several times that. But eventually you will have to grab the wheel or the driver-assist aids turn off.
  • Active Lane Change Assist is a radar- and camera-based system that semi-automates lane changes. The flips the directional signal, the car waits two seconds to make sure the lane is really clear, and then moves the car left or right.
  • Active Brake Assist with cross-traffic function similarly has “extended speed thresholds” for detecting cars and pedestrians. The car brakes if it detects crossing traffic that you aren’t slowing for, and will initiate advance braking if it senses you’ve come upon the tail end of a traffic jam, where there’s no room to maneuver. Mercedes says the car will (should) avoid accidents up to 43 mph (70 kph) and will mitigate the severity of collisions if you’re going faster.
  • 2017-Mercedes-Benz-E-Class-Interior-6-300x165.jpg
    Evasive Steering Assist
    works with pedestrian detection to help the driver steer around a pedestrian and then apply steering wheel torque (force) to recover and get the straightened out afterwards.
  • Active Blind Spot Assist (blind spot detection), in addition to highway speed warning, can warn of possible city-speed lateral (side) collisions.
  • Pre-Safe Sound emits a sharp “interference signal” through the sound system if it detects an impending collision. It triggers a natural reflex that prepares the occupants for the sound of the collision if it happens.
  • Remote Parking Pilot pulls the car out of a garage or parking space using a smartphone app, with the driver outside. This lets big cars park in narrow spaces and tiny garages.
  • Car-to-X communication is the first car with integral car-to-X, meaning car-to-anything communications where a connected car ahead helps cars behind “see” around corners or through obstacles. The car is both a receiver and transmitter. As with fax machines, one is useless, two starts the possibilities, and millions make it useful.
 
  • Informative
Reactions: scottf200