SciTransfer
SoliDAIR · Project

Trustworthy AI and Robotics for Automated Quality Control in Production Lines

manufacturingPilotedTRL 6

Imagine a factory where robots don't just move things, but actually 'see' and 'feel' if a product is perfect. Most companies are afraid to use AI because they don't know when it might make a mistake. This work creates a reliable safety net so machines can handle inspections and testing without human supervision.

By the numbers
4
industry use cases for demonstration
12
consortium partners
67%
industry ratio in consortium
The business problem

What needed solving

Manufacturing companies avoid AI and robotics because they lack trust in the safety and reliability of these systems. This leads to inefficient manual quality checks and rigid production lines.

The solution

What was built

A set of generic modules for Visual AI, sensor-based AI, and collaborative robotics, currently being deployed in 4 industrial use cases.

Audience

Who needs this

High-volume manufacturing plantsQuality Assurance managers in electronicsAutomotive component suppliersRobotics system integrators
Business applications

Who can put this to work

Automotive
enterprise
Target: High-volume parts manufacturer

If you are a parts manufacturer dealing with slow manual visual checks — this project developed AI-driven visual inspection that improves process efficiency and flexibility. It ensures high-rate production remains accurate without increasing staff workload.

Electronics
mid-size
Target: Circuit board assembly plant

If you are an assembly plant dealing with unpredictable quality drops — this project developed predictive quality control using sensor data. This allows you to fix process errors before they result in wasted materials.

Consumer Goods
SME
Target: Flexible packaging producer

If you are a producer dealing with frequent line changeovers — this project developed collaborative robots enhanced with AI. This makes the transition between different product types faster and more reliable.

Frequently asked

Quick answers

How much does the implementation cost?

Based on available project data, specific pricing is not provided, but the project focuses on creating solutions that are cost-efficient to develop and replicate.

Can this be scaled to a full industrial plant?

Yes, the project is demonstrating these tools in 4 industry use cases within real production environments to prove they work at scale.

Who owns the IP and how is it licensed?

Based on available project data, licensing details are not specified, but the methods are designed to be easily adaptable and replicable for companies outside the consortium.

How is the AI integrated into existing hardware?

The project uses a combination of visual AI, sensor-based AI, and collaborative robotics to plug into existing manufacturing processes.

When will the results be available for commercial use?

The project period runs until 2026-09-30, with implementation and deployment currently underway in WP3.

Consortium

Who built it

The project is heavily industry-driven, with a 67% industry ratio consisting of 8 companies (including 4 SMEs) and 4 research entities. This high concentration of commercial partners across 6 countries suggests the results are grounded in practical market needs rather than theoretical research.

How to reach the team

Contact Fraunhofer Gesellschaft zur Förderung der Angewandten Forschung EV

Next steps

Talk to the team behind this work.

Contact us to connect with the SoliDAIR consortium for pilot implementation.

More in Manufacturing & Industry 4.0
See all Manufacturing & Industry 4.0 projects