Digital transformation in manufacturing and process industry
Research and Development under Contract
Computer Vision, Interaction and Graphics
Artificial intelligence, machine learning, and decision support systems
Computer vision and image processing
FRAMEWORK
The impact of Industry 4.0 goes beyond simple digitalization, going through a much more complex form of innovation based on the combination of multiple technologies, which will force companies to rethink the way they manage their businesses and processes, how they position themselves in the supply chain value, how they think about developing new products and introducing them to the market, adjusting marketing and distribution actions.
The consortium intends to invest in the research and development of a solution that encompasses different technologies to ensure the final visual inspection of parts for the automotive industry. The parts to inspect are the piston rings, often also called segments, which are components of internal combustion engines with a circular shape, elastic and with high expansion force.
PROPOSED SOLUTIONS
An inspection system that combines efforts in different technological areas, from hardware (robotic systems, high-resolution image acquisition systems) to software (deep learning, image processing, visual analytics, etc.), aims to detect and classify defects in piston rings used in the automotive sector.
As secondary objectives, the following subsystems will be developed:
- Flexible automated system that allows the handling of parts of variable dimensions
- Vision and detection system for nonconformities corresponding to different types of rings
- Software for image processing and analysis uses artificial intelligence techniques, such as deep learning, to detect various types of defects.
- Automatic quality system, with the development of a database to store data (Big Data) resulting from visual inspection
CCG/ZGDV CONTRIBUTION
The interactive system developed for the Piston Rings intelligent inspection Machine (PRiiME) represents a support tool for quality systems in the automotive industry. It results in safer and more efficient data management through visual and analytical resources that allow inferences about system performance over time.
CCG/ZGDV contributes to the project through the CVIG department.
A secure communication layer was developed to facilitate monitoring of the evolution of an automated quality system. The underlying architecture is composed of three open-source technological elements:
- Database (Postgres): to store and manage data from the automated quality system
- Backend (Django): data security, services and data processing
- Frontend (VueJs): Interactive interface for those that allows visualization and interaction with system data