In the publication "Connecting conceptual models using Relational Reference Attribute Grammars" (⚠ DOI pending), a use case involving a simulated robot arm and two different models conencted to it was shown.
In the publication [*"Connecting conceptual models using Relational Reference Attribute Grammars"*](https://doi.org/10.1145/3417990.3421437)(⚠ DOI still pending), a use case involving a simulated robot arm and two different models conencted to it was shown.
One model was used to ensure a low speed of the robot when within a safety zone (purple boxes in the picture below), and the other model executes a workflow to control the robot.



This paper was presented on October, 16h during the [MPM4CPS workshop](https://msdl.uantwerpen.be/conferences/MPM4CPS/2020/) within the [MODELS 2020 conference](https://conf.researchr.org/home/models-2020). For more information, see the [presented slides](material/mpm4cps-slides.pdf) or the [a recording of the session](https://youtu.be/Hgc1qFfmr44?t=1220).
The repository with the used source code can be found at: <https://git-st.inf.tu-dresden.de/ceti/ros/mpm4cps2020>
The usage is dockerized, so starting the application involves the commands listed below.
The usage is dockerized, so starting the application only involves the commands listed below.
As ROS takes some time to start up, it is best to use separate terminals for the three applications (ROS, Safety-Model, Goal-Model):
```bash
...
...
@@ -16,13 +20,13 @@ As ROS takes some time to start up, it is best to use separate terminals for the
./prepare-docker-compose.sh
docker-compose up -d mosquitto # Starts the MQTT broker