AQUILA

Automatic QUality Inspection using deep Learning and Answer set programming

NEED for AI:

The need for AI arises from the current manual process of quality inspection, which is time-consuming and prone to errors. Reducing inspection time and errors leads to cost reduction and higher revenues for manufacturing companies

AI REGIO SOLUTION:

Using computer vision techniques and AI to automate the compliance assessment of an electronic product. This involves comparing the design of the product in a CAD file with a real artifact, identifying the basic components in the real artifact using a trained neural network for object detection, and then comparing the identified components with the expected components in the CAD file using a logical program. The output is a compliance report that indicates if all the expected basic components in the CAD are present in the real implementation and whether they are in the correct position.

EXPECTED BENEFITS:

AQUILA system will increase digitalisation level of QI processes and reduce QI time by about 60%
enhancement of Deep Learning and ASP know-how
strengthens relationship with manufacturing players
results can be used in workplace security and AQUILA can acquire at least 5 QI and 3 workplace security prospects

Revelis will enhance its quality inspection process for industrial products, particularly electrical panels, by developing an AI-based system called AQUILA. By combining computer vision deep learning techniques with automatic reasoning and Answer Set Programming, the system aims to achieve a 30% increase in defect detection precision, resulting in cost reduction and higher revenues. The project is aligned with the Smart Manufacturing approach and will be implemented and tested within a total of 8 months.