There was a horrible fatal accident with that failure. And it was proven so there was a judgement.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Model Y Driving with FSD Accident
- Thread starter wildli
- Start date
While it is likely the OP's wife pressed the wrong pedal in that situation, but I also have experienced where FSDb struggles. In this location, where it is a big wide right turn with an island, FSD behaves differently in each release - fails in the middle, turns really slow making it dangerous, shoot out to the right most lane, turns and then speed up fast, handles it perfectly. Today, I only let it drives by itself if the other direction has no traffic or in red. I have a different experience at another location - without the red lights but only a yield sign, FSDb just drives through it without even slowing down, the merged road's traffic is 60+ mph and the view on the left side is blocked by objects until the car is at the fork. I never use it again at that junction. If I were driving at the OP's location, I will take over and make the turn myself before the turn.
sleepydoc
Well-Known Member
Yes, that’s the point. Toyota had to pay but when some independent investigators (who were not paid by Toyota or any of the attorneys) took a look they found the only rational conclusion was that it was driver error.
Look sharing this is interesting, but truth be told your wife and you are at fault, “fsd” is a marketing gimmick and is not above level 2 autonomy, meaning you need to be aware and ready to react. Blaming the vehicle seems silly here, plurality of the objective posters here are not in your court, so I’d imagine a traffic judge is unlikely to be either…not to mention some simple facts such as your wife clearly disregarded the passenger not fastened annoying warning for your mother in law, and she hit the accelerator incorrectly. Very high probability most other drivers with fsd experience in this situation would take control knowing that this thing gets confused frequently.
Your wife should admit fault in court, likely eat a fine and some defensive driving and points on her license and move on
Your wife should admit fault in court, likely eat a fine and some defensive driving and points on her license and move on
Last edited:
sleepydoc
Well-Known Member
I don’t recall - there was a lot of hysteria at the time. Juries are not rational, they’re emotional and often not that smart so Toyota was kind of screwed, regardless. The podcast I listened to was recent and several years removed from the trial so it wouldn’t have had any effect on it, it was just very enlightening.
sleepydoc
Well-Known Member
For those interested, I found the podcast that looked at braking vs unintended acceleration:
All,
Tesla insurance cames a total loss results but the vaule is only $45000.
Question: if we accept the total loss, can the FSD license be transferre to a new Model Y we are planning to buy? Not sue how the policy goes.
Are there any tips to negotiate with the Insurance for a higher value? Our model Y is about 29000 miles, 7 seat, with a tow, and FSD options. Cannot believe the value is only about $45000
Tesla insurance cames a total loss results but the vaule is only $45000.
Question: if we accept the total loss, can the FSD license be transferre to a new Model Y we are planning to buy? Not sue how the policy goes.
Are there any tips to negotiate with the Insurance for a higher value? Our model Y is about 29000 miles, 7 seat, with a tow, and FSD options. Cannot believe the value is only about $45000
No.
A brand new one without FSD is only $49k before the $7,500 tax credit. So only $54k even with FSD after the credit.
What’s retail and wholesale pricing for your car now? It’s not the price you paid that matters, but what the market dictates. You may find that it’s pretty close to what you’re being offered.
hybridbear
Active Member
ucmndd
Well-Known Member
My experience in software development is not related to Tesla but since you asked, I will elaborate a bit: I wrote embedded software that are stored inside the hard disk flash memory. This software control all aspects of the Hard Disk. As more features get added, more subtle issues are introduced, mostly due to unexpected events. Testing are enhanced to cover those events and this is quite an endless loop. But at some phase, the products passed all internal tests and customer tests. Everyone is happy until a newly but very rare event occurs exhibiting a new problem. New version of software , new test, etc... Been in this HDD industry for 20+ years and this cycle never ends.
FSD is a very complicated software IMHO, it's mistakes could be deadly, so I personally do not let it drive my car.
I will create another thread to highlights some weird behavior of TACC and Auto Steer. Don't take me wrong, I love driving Tesla and want to see its software get better and better in assisting the driver's.
sleepydoc
Well-Known Member
you give a good example of complex software design (even if HDD controller firmware is significantly simpler than FSD) and how difficult it can be to detect, find and quash bugs. I'm quite certain that we've all seen examples of that with the various FSD releases. An attempt to improve one aspect of the code results in another unintended consequence. Such is the case with complex systems.
I'd still disagree with your take on autonomous driving, though, as it fails to recognize that humans have never worked right, either. In HDD parlance, you need to look at MTBF. The failure rate doesn't need to be zero for it to be better.
