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

2023.6.x

This site may earn commission on affiliate links.
Nothing is learned from individual cars. Its all fleet aka crowd sourced. When it first came out, called deep rain, they did say that they took inputs from the cars if you hit stalk to override but it’s been a while since I have seen any evidence of data from our car being download for any aspect. In early days it was a common occurrence, often after most drives. I guess Tesla have more data sources than they need now.
The wiper response model is trained en-masse using data from the fleet, as you said. The inference side of it is, of course, vehicle specific and runs on the vehicle's computational power. That inference result decides the wiper response to real-time vehicle data inputs.

It's therefore perfectly possible for two vehicles to have sufficient disparity between those inputs to infer two completely different results from the same wiper response model.
 
  • Like
Reactions: KL0809
The wiper response model is trained en-masse using data from the fleet, as you said. The inference side of it is, of course, vehicle specific and runs on the vehicle's computational power. That inference result decides the wiper response to real-time vehicle data inputs.

It's therefore perfectly possible for two vehicles to have sufficient disparity between those inputs to infer two completely different results from the same wiper response model.
it should be soooooooo easy, because all these neural networks, ai, machine learning and all other buzz words tesla is enjoying.

auto wipers enabled. if driver presses the stalk button more often than auto wipers wipe - then increase wiping rate.

should be easy.
 
it should be soooooooo easy, because all these neural networks, ai, machine learning and all other buzz words tesla is enjoying.

auto wipers enabled. if driver presses the stalk button more often than auto wipers wipe - then increase wiping rate.

should be easy.
Haven’t done a controlled experiment but the autowipers wiping rate if set in auto is vehicle specific - different cars do it differently. Not sure Tesla’s vehicle specific data tend to manage this but not hard to do a bit of experiment with so many vehicles in Cambridge. Any volunteers - we can test over the weekend - it’s 70% rain on Saturday a drive on long road or babraham road should provide some data!
 
Apologies for the OCD screen cleaners like me (I'll blame the wife)..

Very quick and simple test - can it see a brick wall and a big bush in broad daylight conditions.

It seems able to do the basics - visual orange/red cue along with "regular" USS beeps. The measurements it's showing are relatively accurate, if a bit conservative for (lack of) forward camera - which I expected.

I'd be interested for other people's tests to see what conditions it struggles in.
I'm planning on retesting at night, and as soon as my rear camera gets obscured by muck.
 

Attachments

  • PXL_20230328_110548995.MP.jpg
    PXL_20230328_110548995.MP.jpg
    400.7 KB · Views: 73
  • PXL_20230328_110510391.MP.jpg
    PXL_20230328_110510391.MP.jpg
    382.7 KB · Views: 42
  • PXL_20230328_110515086.MP.jpg
    PXL_20230328_110515086.MP.jpg
    389.1 KB · Views: 39
  • PXL_20230328_110544920.MP.jpg
    PXL_20230328_110544920.MP.jpg
    411.3 KB · Views: 44
  • PXL_20230328_110523862.MP.jpg
    PXL_20230328_110523862.MP.jpg
    388.8 KB · Views: 40