Decoded Vehicles for a Wear and Tear Model

I want to build a valuation model that accounts for wear and tear on the car as depicted by different events from the Autopi recorded trips relative to a general population of the same type of vehicle. There are many things I’m likely going to ask about over the course of this project, but I’m early on in the process.

The number one thing I need is a sense of how many vehicles are decoded such that I can get reliable data. I’m thinking things like RPM, braking behavior and so on. Frankly, I don’t know where to look to see a complete DBC file for any vehicle but my own. Unfortunately, my car is a 2019+ mazda3 and really only has movement PIDs decoded, so I can’t even use the car explorer library to get a good sense of what a largely decoded vehicle can give me in terms of data.

Any ideas on how to get a good overview on the decoded vehicle population?

Good day Fallen,

This sounds like an amazing project. Being able to value your car based on its current conditions or simply predicting wear and tear would be a really good addition to the AutoPi set of tools.

I have done a bit of research and was able to find a small list of DBC files for Mazda cars here. This will further help you put in perspective the population of decoded vehicles related to your 2019+ Mazda3 car. However, keep in mind that a lot of these databases are curated by the automotive community and not by car manufacturers themselves. Therefore, a lot of the files might be incomplete.

Finally, if you want to expand your development into other car makes, you can explore a more extensive list of file from the comaai open source project here.

Happy hacking!


Best Regards,
Dragos

1 Like