A general-purpose decentralized yet coordinated AI platform
| Reference number | |
| Coordinator | Savantic AB |
| Funding from Vinnova | SEK 500 000 |
| Project duration | November 2020 - August 2021 |
| Status | Completed |
| Venture | AI - Competence, ability and application |
| Call | Decentralized AI Systems |
Important results from the project
The purpose of the project was to design, develop and evaluate a privacy-preserving AI solution for inspection of PCB components for SAAB. Our MVP has been developed and evaluated for "PCB non-wetting" and voids. Specifically, our software has been built with containers, Node.js, Python scripts and TensorFlow. The client software has been developed as a Windows 10 UWP program and LinkAI holds ongoing positive discussions with new potential customers in SE, NO and the EU.
Expected long term effects
Our MVP of a privacy-preserving AI-driven PCB defect inspection solution, consisting of both a client and server software/hardware, has been designed, developed, evaluated and presented to SAAB. In the project, we have been able to show the direct benefits of a machine learning (ML) approach when it comes to the classifications of defects on PCB. There are some ongoing business activities to be carried out after the end of the project.
Approach and implementation
The project has been carried out with regular project meetings and demonstrations where all parties have contributed. LinkAI was responsible for the design and implementation of the platform and AI integration, Savantic was the coordinator for the project and developed code for ML and SAAB provided relevant use cases and gave feedback on the design and results on the platform. The project resulted in a software that shows the function of the ML models that can be installed and tested by the user.