Tristan Profile picture
10 Apr, 11 tweets, 2 min read
Got a sample of the Tesla Insurance telemetry data. The insurance records are on a per drive basis. Here's the fields:

* Unique Drive ID
* Record Version
* Car Firmware Version
* Driver Profile Name
* Start / End Time
* Drive Duration
* Start / End Odometer

(1/2)
* # of Autopilot Strikeouts
* # of Forward Collision Warnings
* # of Lane Departure Warnings
* # of ABS activations (All & User)
* Time spent within 1s of car in front
* Time spent within 3s of car in front
* Acceleration Variance
* Service Mode
* Delivered

(2/2)
There's lot of basic stuff which insurance companies can get via companion apps/dongles but there's a lot of deep insights into driver behavior which Tesla can get but others cannot.

I bet a lot of insurance companies would love to get their hands on this kind of data
There's a lot of strong signals on bad driving behavior. From this data it's easy to pull out behaviors like:

* tail gating - time to car in front, collision warnings, ABS activations
* inattention - autopilot strikeouts, warnings
* heavy throttle usage - acceleration variance
It's interesting that they track rapid acceleration considering that Tesla sells performance cars known for fast acceleration.
The driver profile name (which is often uniquely identifying) allows Tesla to know the # of different drivers as well as differentiate different driver behavior.

Notably there's no VIN logged in each record but that would be easy for Tesla to include on the server side.
This is just the initial data. The version field means that Tesla can easily add new data to be tracked later.
The service mode and delivered fields are presumably there to filter out drives not done by the owners. Though I wonder if Tesla can also use it to detect joy riding, etc by service centers. If they were just filtering they could do that before logging the drive.
There's two ABS fields one user and one not. Presumably this is to distinguish rapid braking due to inattention vs ABS kicking in from ice or snow though it's hard to say for certain
I also didn't see anything in the UI to indicate that Tesla is collecting this data -- but I may have missed it somewhere. Certainly wasn't any "opt-in" dialog like the driver monitoring
The forward collision warnings is very interesting since the user can adjust how sensitive the activation is. That seems like it would make it less strong of a signal since people who commonly tailgate would turn it off

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Tristan

Tristan Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @rice_fry

12 Apr
We recently got some insight into how Tesla is going to replace radar in the recent firmware updates + some nifty ML model techniques

⬇️ Thread
From the binaries we can see that they've added velocity and acceleration outputs. These predictions in addition to the existing xyz outputs give much of the same information that radar traditionally provides
(distance + velocity + acceleration).
For autosteer on city streets, you need to know the velocity and acceleration of cars in all directions but radar is only pointing forward. If it's accurate enough to make a left turn, radar is probably unnecessary for the most part.
Read 15 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!