MoDeS3 – Model-based Demonstrator for Smart and Safe Systems

June 8, 2016    

We present MoDeS3, a Model-based Demonstrator for Smart and Safe Systems,

a project which won the 3rd prize at the 2016 Eclipse Open IoT Challenge 2.0. The fundamental goal of MoDeS3 was to create a demonstrator for educational purposes that showcases how industry standard IoT solutions from the Eclipse ecosystem can be integrated and combined with modeling techniques to develop smart and safe systems.

Specifically, MoDeS3 is built on a fully functional model railway case study (featuring Beagleboards, Raspberry Pis, sensors, actuators, cameras, a LEGO robot, and lots of custom electronics), where classical safety critical engineering principles (such as statechart modeling and verifcation) are combined with modern sensing (computer vision), communication (MQTT) and complex event processing (VIATRA) technologies to showcase how the smart and safe systems of the (near) future might look like – from an engineering perspective.

The highlights of the talk are:

  • We describe the state-of-the-art modeling toolchain we constructed to build MoDeS3. The toolchain is centered around Yakindu Statecharts, a popular modeling platform based on Eclipse technologies, which we augmented with validation and code generation capabilities using Eclipse’s VIATRA model query and transformation framework.
  • We overview the runtime architecture of the system, which is built using popular IoT components such as MQTT messaging (Mosquitto, Paho), a web-based UI, sensors and actuators connected to Beagleboards and Raspberry Pis, a LEGO robot, and computer vision using OpenCV.
  • We discuss how classical safety critical engineering principles (such as modeling and verifications) can be combined with modern IoT technologies to address the challenges of building complex systems of mixed criticality. Specifically, we will highlight how models can not only be used at design time, but also at runtime, to support e.g. system monitoring and fault detection with complex event processing.
  • We will show several videos and small demos that showcase the highlights and capabilities of the system and its components.

The work presented is developed in collaboration between the Fault Tolerant Systems Research Group of the Budapest University Technology and Economics (BUTE), the MTA-BME Research Group on Cyberphysical Systems (a joint research group of BUTE and MTA, the Hungarian Academy of Sciences), IncQuery Labs Ltd. and QuanOpt Ltd.