<!- The project is part of a broader strategy for modernizing production lines aimed at exploiting the benefits achievable with the application of the Industry 4.0 paradigm. Machine learning and data sensors’ acquisition techniques were used to develop a distributed control system which includes an advanced mechanical modeling. The project was jointly developed with the University of Pavia and Omron, the latter providing the preview of its iForest technology. The results have been very encouraging and the solution is being applied for revamping various production lines

Requirement: AI algorithm application, ML strategy development in production environment.
Partner: Omron – Enoplastic – Università degli studi di Pavia
Milestones: 2018, manufacturing of the first machine with AI-based controller; 2019: tests; 2020: first line in full production; 2021-22: 20 units scheduled. ->


[Dark Army]

<!- Quality control system for prints and accessories developed for ENOAL capsule forming machine.
Video system for checking:
– Presence
– Quality
– Registration of prints made with hot-foil technology and accessories made in the machine.
The project involves the use of a GigE Poe type DALSA Vision color camera Genie Nano C4024.
The software allows the image processing through MVTech Merlic, it allows the application of filters, the creation of ROI and the attribution of scores to the various acquisitions according to “customizable” criteria. The continuous storage of data collected by the video system will be the basis for the development of deep learning strategies that will allow the machine to autonomously manage the process based on the information it has acquired.

Requirements: Visual recognition systems, machine learning, advanced robotics, multidisciplinary machine
Partner: Enoplastic-Correlance
Milestones: 2020: kick-off. ->


[Deep Sight]

<!- The project aims to develop a tracking system for moving elements that can be applied to machinery of different types from automotive to industrial automation with the use of low-cost hardware.

Requirement: Application AI algorithms, neural network, Nvidia HW, vision systems
Partner: TODEMA
Milestones: kick-off: 2020, HW and SW development and tests: 2020 ->


[Psyco Zaku]

<!- This project concerns the use of a neural interface for the command and control of automatic devices. In particular, the objective of the research, currently underway, is to outline the horizon by identifying the technical solutions and algorithms present on the market or in mature scientific literature suitable for industrial applications. Through the neural interface it will be possible to extend the range of use of driving simulators not only for the motorsport sector but also for autonomous driving and for the functional rehabilitation of a patient’s abilities following trauma. The experience of using the neural interface in the simulation system will provide the operational contours around which to plan the use of this technology in the industrial field, both for the support of HMI platforms for the control and monitoring of production plants and for platforms of augmented reality. The use of an OPENBCI EEG Electrode Cap type helmet with 16 channels will allow the analysis of the subject’s neural electromagnetic impulses that will be interpreted by the openBCI software connected to the cyton board and daisy module and analyzed directly in Matlab / Simulink by means of tools like BCIlab.

Requirement: Dedicated hardware, man-machine interface on a neural basis, advanced robotics, biomechanics
Partner: University of Pavia – ApiTech
Milestones: started in 2019, in progress ->