Jump to content

ABRP Blog

  • entries
    22
  • comments
    153
  • views
    365,506

Contributors to this blog

Helping ABRP - Improving Car Models

We have three levels of car model confidence which we use to label the cars in the planner:

  1. Alpha - Modeled manually by us, these are generally closely matched to the EPA and WLTP range provided by the manufacturer, but won't be a perfect representation of the car.
  2. Beta - Modeled based on some real-world data.
  3. Release (no label) - Modeled based entirely on real-world data.  

Live Data

To move a model out of Alpha, we need real-world data, and that's where you come in!  The best way to get that real-world data is to use a Live Data connection.  The first step there is to check your car against our list of Live Data Support and see if it's already supported.  If so, great! Set up Live Data, and once we have enough data we'll promote the model from Alpha.  Of course, setting up Live Data also has a ton of benefits for you as a driver as well:

  1. We do live calibration of your personal car and account for things like extra passengers, cargo boxes, bikes, and weather automatically
  2. You get a live graph and arrival predictions of state of charge while in Driving Mode
  3. You get completion estimates while charging and can see charging status

And of course, if we don't currently support your car, but you know how to get data from your car we provide an API to connect Live Data with us.  Send us an email at info@abetterrouteplanner.com if you'd like more information about this, and our API Docs are available on our Postman.

Testing Procedure

If we don't have Live Data support for your car, the other option is using the Trip Odometer.  We've prepared a test procedure document you can print out and fill in to provide this data.  See the attached "Manual Car Calibration" document.

The goal here is to provide a series of data points at a variety of speeds to estimate the efficiency across all speeds.  With many tests on different copies of the same car, we can produce a reliable model.

Manual Car Calibration.pdf



2 Comments


Recommended Comments

Jason - if following the manual test procedure, wouldn’t it also be useful to report the weather conditions at the time the test was conducted?

Also, could you perhaps outline which car models you don’t really need any manually collected data for - i.e., because you’ve already got sufficient data and/or there are enough people providing automated live data? For example, would it still be beneficial to provide manually collected i3 data even though automated live data collection is now possible?

Thanks

Edited by Bill N

Share this comment


Link to comment

Also, it might be worth noting that if people just set their cruise control at a given speed (eg 60 mph), then the actual speed they are travelling at could vary by a few mph depending on the car model due to different levels of speedometer accuracy between manufacturers.

So it might be worth advising people doing the manual test procedure to set their cruise speed based off a GPS speedometer (eg Waze) in order to eliminate that variance.

Share this comment


Link to comment
Guest
Add a comment...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Contact Us

Bo - Lead Developer and Tesla owner: bo@abetterrouteplanner.com

Jason - New Car Models, Developer and Bolt owner : jason@abetterrouteplanner.com

Idreams - Forums Administrator, Forums Developer and Tesla owner : idreams@abetterrouteplanner.com

×
×
  • Create New...