Wol747
Active Member
Very inconsistent. Today I was in an 80 Km/Hr area and came against a road work 40Km/Hr sign: the car recognised that but just carried on with it after the zone until it saw another 80 sign miles down the road.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
65 mph Speed Limit Signs "invisible" to Tesla
- Thread starter SSonnentag
- Start date
I think the car uses some blend of speed limit from map data and reading speed from sign.
The car obviously can read a clear construction zone 55 mph sign, yet it is ignored. I expect the system was misinterpreting signs like truck speed limit so they chose to ignore camera speed limit in some situations.
The car obviously can read a clear construction zone 55 mph sign, yet it is ignored. I expect the system was misinterpreting signs like truck speed limit so they chose to ignore camera speed limit in some situations.
whitex
Well-Known Member
When it comes to reading signs, AP2 has been way behind AP1 for years. Elon keeps thinking "who knew speed signs were that complicated!?! Oh yea, those MobileEye guys knew"
Tesla's sign reading problems is a failure of deep learning.
So is the issue of cars missing from the display when stopped at a light. Getting most of the desired functionality fairly quickly is a characteristic of deep learning. So is hitting a brick wall regardless of the quantity of additional training. Tesla has apparently not achieved the advances in deep learning that they needed to achieve FSD.
So is the issue of cars missing from the display when stopped at a light. Getting most of the desired functionality fairly quickly is a characteristic of deep learning. So is hitting a brick wall regardless of the quantity of additional training. Tesla has apparently not achieved the advances in deep learning that they needed to achieve FSD.
DevilDoc8542
Member
What’s wild is driving down I-35E towards Dallas, we have toll roads in between the north and southbound lanes. If I am in the left hand lane closest to the center toll roads, the car will recognize the 75mph signs and then if you move towards the center and right hand lanes, the car will pick up either the 65 and/or 70 mph signs.
M3BlueGeorgia
Active Member
Apparently more simpler than that.
It was rumored that Mobileye had IP or patents, which have either now expired or Telsa found a way around.
Yeah...there’s a VERY long tail on this thing that the machine learning will not get. I have close to zero confidence that full self driving without requiring a drivers hands on the wheel is coming anytime soon.
The car has to handle 100% of situations right 100% of the time. Too many crazy scenarios and variables in the real world for the computer to understand them all.
Humans don't. Autonomous vehicles have to be measurably better than humans almost all of the time and then try to handle the other situations as safely as possible.
When Tesla started their FSD other sophisticated companies were very optimistic about the reach of machine learning. The voice assistants are the best example. Even Apple was very optimistic about Siri in 2016. All these companies hit the wall where more data doesn't improve results.
Tesla experienced the unfounded optimism in the now famous "3 months maybe 6".
So what Musk is today claiming is that Tesla have broken through this machine learning barrier. Yet he can't know, because they have not proven a large improvement in perception.
Today it is hard to be nearly 100% certain to simply determine if a human is present in a video image. Security camera systems struggle with this seemingly simple functionality. If Tesla has made a breakthrough a great many products can be improved with better machine learning.
The difference there is the humans take the liability and are held responsible. Is going to accept liability and responsibility for every mistake their cars make? Are regulators going to ALLOW Tesla to take that liability if they’re willing? You can understand the difference here, right?
Yep. I work for a large publicly traded software company. We employ a number of PhD data scientists that specialize in machine learning, image recognition, spacial analysis, etc...
Like I said, the number of variables it too great when considering situations you encounter while driving in the real world. Getting it right 90% of the time isn’t going to cut it...not even close.
TicoMcC
Member
I ran into this case just yesterday for the first time. Sign said "35 mph limit ends". Car stayed at 35 mph. The speed limit was then back to 50 mph.
Similar threads
