Development of the physical twin of a Fischertechnik-based factory

Context:

The MBDO project (Model Based DevOps) aims to provides methodologies and tooling to implement digital twins.

Goals of digital twins contains: monitoring, deviation detection, failure detection, simulation, planning, automatic or assisted reconfiguration, …

As a digital twins should represent a physical twin,
the DiverSE team has acquired a set of Fischertechnik modules that will act as the physical part.

The modules creates a Training Factory composed of Vacuum Gripper Robots, Automated HighBayWarehouse, Converyor, MultiProcessing Station, Sorting Line.

The actuators and sensors of the modules are driven by several RasberryPi with IO bus support.

Objectives:

Build a realistic support for the factory machines (ie. Fischertechnik modules) in order to implement various scenarios on different factory configurations.

Task list: (The task list will be adapted with respect to the actual internship duration)

• Improve machine controller. The controller acts at the IO level of the engines and sensors and provide a socket based protocol (Ie. improve machine abstraction on top of actuators and sensors, add more tests (with simulation/mock support))
• Implement a set of orchestration scenarios (Ie. implement tools working on top of the protocol to perform high level scenarios on differents factory layouts)
• Implement a shopping web application (Ie. implement an orchestration coupled with a web server that provide an end user reservation/buy/)
• Add MQTT message service to both the controller and orchestation levels. This service will serve as the base interface for monitoring and synchronizing the physical and digital twins.

Candidate

Send an email to Didier Vojtisek