NetMeeting Default Template

Pre-meeting stakeholder’s reports

HDM

• Is developing the Into-cps cloud app.
• Will focus on writing the extended version of the accompanying paper.
• Waiting for details on the submission of an extended version for the accepted paper at co-sim workshop.

Nick

• Extended the dynamic semantic testing to include a combinatorial test suite for Model Exchange in addition to Cosimulation.
• Adding abstract function calls to the API functions to represent the “implementation” of the FMU, which will enable the semantic model to realise any particular FMU’s behaviour - it’s about half done so far.
• Updated the Overture workshop paper for Carl, to include updated results and a link to the online version of the checker.

Claudio

• Won’t be able to attend the meeting.
• Working at CMU on an FMI exporter for ModelPlex.
• Going to meet with Stefan some time this week to discuss other research topics.

Casper

• Working on ESA case. Critical Design Review in late November, 2019.
• FMI Utilities website online. Could use some prettying :)
• Went to Excelgo workshop, might be future cooperation for visualising systems
• Worked on docker-based FMU cross compiler out of necessity.

Jim Woodcock

• Apologies for not being able to make the net call: I am at the RoboSoft meeting in London.

Corneliu

• Continued to investigate the PTC ThingWorx platform and training material. Working with the project group to connect a PLC with ThingWorx and exchange data, etc.

PGL

• Presented about digital twins 3 times over the last two weeks
• Got AU involved in 4 new EU applications related to digital twins with different roles (the last 3 will be submitted tommorow)
• Agreement for a desktop version of Robotti has been established
• Progress is reported from Matthias on the indoor positioning system
• Saw demo of Siemens Amesim
• Arranged meeting with Xcelgo about potential collaboration on digital twin technology (takes place next week)

Carlos

• On going Discrete Model of a manufacture process with VDM++

Mathias

• Working on the local projection to determine the objects position on the track, together with its rotation.
• Can detect the circle’ position
• Working on identifying the id for each circle, to mactch with the digital twin

Sub-projects status

Into-cps cloud app (HDM)

• Found some bugs in the COE assignment algorithm.
• Improving backend testing.
• 50 user version in progress.

Maestro V2 DT enabled (CT)

• Nothing new.

Checker of FMUS (NB)

• Only a few changes to the static semantic model - most (transparent) changes to make the dynamic semantic modelling easier.
• Now that the dynamic sematics is close to “working”, I want to move over to static semantic for FMI 3.0.

Tolerance Language (JW)

• I’ve been discussing the tolerance language ideas with Marcel Verhoef (ESA).
• Marcel recalled a paper that he wrote with John Fitz and Peter: Validation Support for Distributed Real-Time Embedded Systems in VDM++.
• This paper defines validation conjectures, which are statements of compliance of observed behaviour against a specification. This is a good approach that we might follow in the tolerance language.
• Spacecraft operations generate large amounts of housekeeping and science data that is downloaded over telemetry links. This is a well understood application area where data streams are defined upfront. This includes their expected ranges and alarms in case of parameters going out of bounds. The approach doesn’t seem particularly sophisticated, but it is supported by a small script language used to define “synthetic parameters”. These are computations and transformations performed on low level parameters in order to derive system state, specific situations, etc. This is all relevant to Digital Twins.
• At ESA, they analyse telemetry streams using big data and machine learning approaches. Their objective is to predict errors before they appear in telemetry to assess slow decay, wear and tear, etc.
• In another piece of work, I am writing a paper on the behaviour of the Aethon TUG autonomous mobile robot. This has noisy infra-red sensors. The robot uses the Ransac algorithm to get a probabilistic guarantee that its motion detection is free from erroneous sensor values. I expect to have a draft of this paper by the end of the year.

Graphical configuration inside App (CL)

Next Meeting

27 November 2019, 1030 CET