Artificial Intelligence approach for Product Inspection through the use of Cobot enabling quality control of Unshaped products Solution

Informations projet

APICUS

N° de convention de subvention: 836165

DOI

10.3030/836165

Projet clôturé

Date de signature de la CE 28 Novembre 2018

Date de début 1 Decembre 2018

Date de fin 31 Mai 2019

Financé au titre de

• INDUSTRIAL LEADERSHIP - Innovation In SMEs

Coût total

€ 71 429,00

Contribution de l’UE

€ 50 000,00

Coordonné par

VIDEO SYSTEMS SRL
Italy

Periodic Reporting for period 1 - APICUS (Artificial Intelligence approach for Product Inspection through the use of Cobot enabling quality control of Unshaped products Solution)

Période du rapport: 2018-12-01 au 2019-05-31

Résumé du contexte et des objectifs généraux du projet

Todays’ market of hollow glass containers manufacturers is facing increasing complexities in performing quality controls at the cold-end part. Customers’ needs impose customisation of products which result into more and more non-standardised and irregular shapes. Current technologies are limited to examine standard shapes and specialized per type of defects causing false positives phenomena, increasing production costs and slowing-down times. Quality of hollow glass containers are paramount to ensure shelf-life properties and safety of products for customers, as they are used as packaging for food and beverages, perfumery and pharmaceutics. There is no technology in the market that performs thorough quality controls checks of all type of defects and that inspects non-standardised shapes.
Video Systems aims to develop an innovative inspection system for the cold-end and the reselection phase, APICUS. The objective is to empower a collaborative robot with artificial intelligence and artificial vision able to handle an article in a human-like way.

Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent

A technical/technological feasibility assessment and a market search have assessed and validated the technical and economic potentiality of the APICUS technology.
From the technical point of view, artificial intelligence and artificial vision algorithms were examined to understand which suit best the purpose of integration with a collaborative robot. Two prototypes were tested in a simulated environment. Results showed high degree of efficiency and defect detection together with a low degree of false positives. Technical improvements were identified to refine the APICUS before market entrance, which will be subject to a phase 2 development of the product.
From the economic point of view, customers’ needs were assessed. These have brought to evidence that beyond the hollow glass containers sector, other market sectors are experiencing the same need of high-efficient and high-reliable automatized technologies for quality controls, such as the food industry to check the quality of food or the mechanical sectors to control high precision components. APICUS’s value proposition lays on highly accurate quality controls for every type of defects, at reduced times and costs.

Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent)

Important advancements compared to the state-of-the-art have been made with the integration of artificial intelligence and vision algorithms into a cobot to enable to handle articles as humans to perform quality checks. These have brought to increase in efficiency and accuracy of defects detection and reduction of false positives and wastes from production lines. The expected results are to refine technically the APICUS prototype to achieve full-envisaged functionalities and desired performances.
The socio-economic implications comes from a revolutionised approach to quality controls that reduces costs and time-to-market for manufacturers, increases the satisfaction and safety of customers, improves job conditions of workers implied in quality checks tasks.

APICUS - Glass on glass tests

6 axis robot Prototype station hollow glass

APICUS - Bottom bottle contaminant identification

APICUS - Shoulder crack identification