KITT4SME live platform
The AI's coming to the shop floor!
This repo contains a work-in-progress reference implementation of the KITT4SME platform.
The what & the how
The what: affordable, tailor-made AI at scale
- Manufacturers on a budget and no IT expertise can still benefit from AI.
- AI developers can focus on delivering business value.
- The platform provider pools resources to slash IT costs.
The how: a service mesh, multi-tenant, cloud architecture to
- assemble AI components from a marketplace into a tailor-made service offering for a factory;
- connect them to the shop floor;
- enable them to store and exchange data in an interoperable, secure, privacy-preserving and scalable way.
Docs
We have way less docs than we'd like to, but we'll keep on adding stuff as we go along. Here's what's available at the moment.
- Architecture. This is the repo where we keep most of the high-level tech docs, including a catalogue of the services the KITT4SME consortium are busy developing—the catalogue doesn't yet include the awesome software the open call winners will develop. Give those docs a read if you're trying to figure out how the whole thing hangs together :-)
- Cluster bootstrap. Read this if you're itching to build your own KITT4SME cloud instance. If you're familiar with cloud computing and micro services architectures, you can probably whip it together in about 20 mins.
- Security Ops. Managing cluster secrets, setting up SSO, and all that jazz.
- Storage. How to manage cluster storage.
- Deploying an app service. How to package an app service for deployment, configure the KITT4SME platform instance and finally go live.
- NGSI persistence. Short explanation of how KITT4SME automatically stashes away your NGSI data and builds time series out of it.
- WP6 pilot deployments. Outlines how the KITT4SME platform got packaged and made available for the WP6 pilot deployments.
Hacking
You're welcome to fork this repo and submit a PR :-) The deployment
directory contains the Kustomize code we use to build the platform
live at kitt4sme.collab-cloud.eu and keep it in sync with this repo.
The code is organised in sub-directories to mirror the layers detailed
in the Cloud instance section of the architecture document.
If you want to develop your own platform app service, you could use our Roughnator as a starting point. Roughnator is a REST service that estimates surface roughness of manufacturing parts. The app receives manufacturing part measurements from Orion (Context Broker), comes up with a roughness forecast and then stashes that away into Orion. The KITT4SME platform automatically builds time series of both input measurements and output roughness estimates, making them available to other services. The Roughnator repo also contains integration and end-to-end tests to verify service functionality in a local Docker environment that mimics the platform live environment where the service ultimately gets deployed. Also, the GitHub repo is configured to run a CI/CD pipeline that publishes Roughnator's Docker image on each software release.
Project management
We use a GitHub project to manage platform development. If you're curious about what we've been up to and where we'll go next, head over to our project site where you can find out about objectives, roadmap and sprints. We've got lots of dev activities going on in KITT4SME from design to coding to writing tech docs, there's plenty to keep us busy. We put together a short intro on how we use the GitHub project to keep tabs on dev activities and get stuff done: