Despite the summer break, partners in the EMPIR project "Metrology for the Factory of the Future" have been very active, and I would like to give you a quick update about the most important recent developments in the project.
Tutorials on application of machine learning to testbed data sets
An initial set of tutorials has been made available online that demonstrate how machine learning methods can be applied to the data sets that have been uploaded to the Zenodo Met4FoF community page. The tutorials are interactive and can be run in the cloud without any installation on the computer necessary.
The next step we are working on is the extension of these notebooks to demonstrate the effect of uncertainties in the input signals.
Smart-up unit demonstrator with EMPIR SmartCom
The smart-up unit is a microcontroller board that can extend digital-only sensors, such as MEMS, and amend their output with traceable time stamps and preprocessing of the measured data. The code for the microcontroller is open-source and available online. We are currently working on a demonstrator jointly with the partner project SmartCom. That demonstrator will use the smart-up unit with a calibrated MEMS accelerometer. Communication between the µC board and a connected laptop is based on OPC-UA implementing the digital SI (D-SI) data structures from SmartCom. Data pre-processing is carried out using methods from PyDynamic.
Multiagent system webinar
Recently, we organised a webinar on the multiagent system developments within the project. With the webinar introducing the MAS and giving a quick tutorial on its use, we intend to initiate a long-term development and collaboration. All future developments within WP2 of Met4FoF will feed into the MAS infrastructure. Other related projects will also make use of this MAS. Hence, we are committing ourselves to its further development..