I predict, expropriating your metaphorical world, that this will happen around the same time I can throw some avocados, limes, an onion, a tomato, and some cilantro into a hopper and have perfect guacamole come out the other end.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
I predict, expropriating your metaphorical world, that this will happen around the same time I can throw some avocados, limes, an onion, a tomato, and some cilantro into a hopper and have perfect guacamole come out the other end.
JohnnyG
Weee!
Hi, I was trying to learn more about the eight camera system that your 2013 Audi A6 used for active lane assist, but I'm having a hard time finding any detail on an Audi system with those specs. Can you share a link, or maybe you have some paperwork at home about it's eight camera active lane assist? I'd love to know more! Thanks!!
Ulmo
Active Member
mmd
Active Member
Hi Folks,
Do you report these to NHTSA as safety concerns? These software glitches can be as dangerous as hardware issues. I don't know if NHTSA wants to poke its nose here, or if it can even do anything, since Tesla is running ahead of the regulations for autonomous cars.
But these seem to be reportable incidents, no? What does the manual say?
Do you report these to NHTSA as safety concerns? These software glitches can be as dangerous as hardware issues. I don't know if NHTSA wants to poke its nose here, or if it can even do anything, since Tesla is running ahead of the regulations for autonomous cars.
But these seem to be reportable incidents, no? What does the manual say?
Compared to the initial versions in early 2017, the recent firmware is much better. When it is working on gentle curves in a freeway it tracks 80% as well as AP1. However, unlike AP1 it can behave very erratically for no apparent reason and literally smash the car into something. On moderate curves it simply doesn't work with any regularity.
Rest assured, someone will die using AP2. It is very unsafe for non-technical users. I won't allow any other drivers in my household to operate it. I don't use it when I have passengers. I use it when alone knowing full well that you have to watch it as if a narcoleptic or drunk has the wheel.
Tesla will not be able to solve this. The big empty spot on the circuit board for the PX2 computer where the EyeQ3 processor was supposed to go speaks volumes. The hardware simply isn't there to crack this nut.
Those of you who read my posts know this is a very sore spot for me, since not only do I feel defrauded by Tesla, but I feel Tesla is gambling with people's lives out of greed.
Rest assured, someone will die using AP2. It is very unsafe for non-technical users. I won't allow any other drivers in my household to operate it. I don't use it when I have passengers. I use it when alone knowing full well that you have to watch it as if a narcoleptic or drunk has the wheel.
Tesla will not be able to solve this. The big empty spot on the circuit board for the PX2 computer where the EyeQ3 processor was supposed to go speaks volumes. The hardware simply isn't there to crack this nut.
Those of you who read my posts know this is a very sore spot for me, since not only do I feel defrauded by Tesla, but I feel Tesla is gambling with people's lives out of greed.
Statistically, autopilot is safer than the average of all human driving. But what happens when you compare Autopilot safety to the safety of only human drivers under the ideal conditions that Autopilot is normally used in? That is, what happens when you compare Autopilot to
human drivers in new cars, with VSC and traction control and anti-lock brakes, on divided highways, within 10mph of the speed limit, with clearly marked lane lines, and with no snow covering the road or very heavy fog/rain obscuring vision? Given the price and technology focus of a Tesla, you should also exclude most teenagers, college students, and the elderly.
Is Autopilot still safer than a human driver driving in those conditions? What about if you include the accidents that occurred seconds after autopilot disengaged itself?
While that statistic is true it is also flawed. Namely it doesn't prove AP is safer by itself. If I was driving myself I would not have hit the median myself on the carpool lane. If I let AP do its own driving without disengagement the car would have hit the median many times over. However with me staying alerted and being aware of AP quirkiness I can correct it and not kill myself. I have to focus on the road and being defensive about when AP will try to kill me. So yes with AP and a more aware driver it does prove to be safer. AP alone? No way.
1. 40% accident rate reduction was with AP1
2. 40% accident rate reduction was observed with Teslas with AP1 vs. Teslas without AP1. But Teslas before AP1 ALSO LACK AEB! I believe this 40% accident rate reduction is based solely on the simultaneous introduction of AEB.
2. 40% accident rate reduction was observed with Teslas with AP1 vs. Teslas without AP1. But Teslas before AP1 ALSO LACK AEB! I believe this 40% accident rate reduction is based solely on the simultaneous introduction of AEB.
What I said is that it used eight cameras, full stop. I didn't say, didn't intend to imply, and don't believe, that all of them were involved in lane-keeping. The car, which was as I recall the Prestige model with the Innovation package, featured a "birds-eye" camera view, to which I think some of the cameras were dedicated. I retained no paperwork when I sold it, so my online research - were I to undertake it - would be no better than yours.
My mention of the Audi having and using eight cameras was mainly a poke at Tesla, which rather pathetically still leaves six (or seven?) of its own eight cameras sitting helplessly by while the AP2 system careens wildly towards other vehicles and fixed objects.
OilSucks
Member
While I'm not happy you experienced this OP, I am happy others have experienced similar events and we can stop this crap talk that Tesla is perfect. I'm not an investor either, just a fan of Tesla and always will be... But AP2 is nowhere close to being on the levels of AP1 and is flat out dangerous with it's random failures. Like you, I had a moment that was very similar and the only 'excuse' I can give it was the sun being right in the camera... But instead of veering me to the right in a violent fashion, I would have also preferred a take over tone or some sort of warning that not everything was on the up and up with regards to the AP system.
OilSucks
Member
The problem is Tesla really should not be advertising either systems (AP1 or AP2) as 'Autopilot' and should be advertised as Driver Assistance tools. The two terms are completely different and I'm afraid the term Autopilot invites a sense of inattention when driving. I get that it's in beta mode, and there are warnings about it, but AP2 really is not on the same level of AP1. For that reason Tesla needs to really help people understand that we are test subjects for it.
As ever the hyperbole kicks in.
No the car did not try to kill me.
Nor will it kill anyone if you drive the car as instructed byt the manufacturer with your hands on the wheel.
If I hadn't been paying attention I might have scraped the front corner is all.
The Autopilot a very good driver assistance system (arguably the best available today - where's the comparative reviews btw) and mostly works well, but it is a bit of a twitchy bugger in the current release that I have; which I notice is already three released builds behind the latest so no doubt it is already improved.
Overly agressive application of steering at speed is the bit I would really like to see them dial out soonest, so it become less of a test of how alert you are (although you should be anyway).
OilSucks
Member
I mean it might be a bit over dramatic but he's not wrong. It's a bit dangerous for a car to have such high levels of unpredictability when driving at 70+mph. The margin of error gets smaller and smaller as the speed increases too. Luckily for me, I had a 'veer off' moment while my hand was on the wheel going 65mph, which is jarring, but I could correct it without losing control. Others might not be so lucky.
That's why Model 3 is the affordable option. It will take A LOT of cars sent on this journey for one to make it all the way there. In any condition. They may have to drive exclusively by night.
All this talk of machine learning, but still being caught out by ancient tar lines...
You're right; they said one car would do it, but didn't say with how many would be trying. Given your observation, I'm going to reverse my metaphor:
The coast to coast unassisted drive in a current-production Tesla will happen when you can pour guacamole into a hopper and have an avocado, a lime, an onion, a tomato, and some cilantro leaves pop out the other end. IOW, when the Second Law of Thermodynamics is repealed.
So guys.. for us model 3 reservation holders... what's your opinion on EAP.. should we get it?
jjh1234
Member
Man, that's a tough call. I rented a couple of AP1 Teslas before buying. I really liked the AP1 on the rentals, even after 1,200 miles. I was looking at some used Teslas at local dealers, one without AP and one with AP1 hardware, but not activated. I really wanted AP, but didn't want to pay $60,000+ for a used MS70D and then have to pay to activate AP1 and have a car that's already kind of out of date, with only 3G, AP1, etc.
When I test drove the demo MS75D AP2 car I ended up buying, it was in April/May and it was still limited to 35 or 45 mph. At that point, I was unaware of all the drama behind the rollout of AP2 and the breakup with Mobileye. I was assured it was almost ready for full speed and would match AP1 right away. If I had been aware of the TMC forums and had read all the posts about the AP2 problems, I probably would not have bought it right away.
Even if you don't use the Autosteer, the TACC is still great on the highway or when stuck in traffic. I'm just not sure it's worth all that money by itself. You have to decide if it's worth it to you. If you plan on keeping the car a long time and are pretty sure you will get it eventually and don't mind it's not perfect yet, get it now. If you are on the fence about really having to have it and might trade up in a year or two, I'd wait to see if it gets a lot better and pay a little more later.
On the other front, I'm way glad I didn't drink the CoolAid and pre-pay for FSD. I suspected it was way in the future even back then and I probably would trade the car before it was ever available. I will also gladly pay the post-sale price if it shows up before then.
Good Luck!
Similar threads
- Article
