search

eclipse-sdv-blueprints / fleet-management

A close to real-life showcase for truck fleet management where trucks run an SDV software stack so that logistics fleet operators can manage apps, data and services for a diverse set of vehicles.
Apache License 2.0
16 stars 10 forks source link

Implementation of COVESA Fleet Management Data Guidelines #43

Open tspreck opened 7 months ago

tspreck commented 7 months ago

It would be great to showcase a few of the recomendations using the new COVESA Fleet Management Data Guidelines (FMDG) instead of the FMS standard. The idea here being that by using the COVESA standards we are more open and transparent for the ecosystem. More background info can be found here for the FMDG

See reference to the current set of guidelines here

Maybe we could create a new feature branch where we can work on this together @tguild @noci0001 or happy to discuss how the working mode could be here as we will want to use geotabs rust Curve algo also to fufill the smart sampling requiremtn of the FMDG

eriksven commented 7 months ago

@tspreck having an implementation of the referenced guidelines would be amazing.

I propose to have multiple steps for this:

  1. have an initial implementation of a forwarder which consumes some of the relevant signals and then applies the Curve algorithm before sending the data to the backend/database
  2. select an initial set of use cases to support and implement this in the forwarder including the generation or recording of test data
  3. implement an efficient way to serialize the data for the transport to the backend/database in alignment with the guidelines

After going through these steps we could have another discussion on how and whether to extend the forwarder to support the full guidelines. For now, we can call the proposed forwarder cv-forwarder or covesa-forwarder.

What do you think?

Regarding the feature branch. I do not think that it is necessary to have dedicated feature branch for this forwarder in the Eclipse repository. Instead we can continue to collaborate through PRs to the main-branch here.

tspreck commented 6 months ago

For topic no 2. Here is a a starting point for the "Fleet Energy Predictor" use-case https://github.com/COVESA/commercial-vehicles/issues/12

image

In the COVESA CV group we will aim to transfer these data points into a "Data campaign" for it

We should be able to find all the datapoints and sampling requirements here for implementation > for more questions this is the best contact source @cgrewer1 & also they have a slidedeck here