A model-driven engineering process for autonomic sensor-actuator networks

Resumen

Cyber-Physical Systems (CPS) are the next generation of embedded ICT systems designed to be aware of the physical environment by using sensor-actuator networks to provide users with a wide range of smart applications and services. Many of these smart applications are possible due to the incorporation of autonomic control loops that implement advanced processing and analysis of historical and real-time data measured by sensors; plan actions according to a set of goals or policies; and execute plans through actuators. The complexity of this kind of systems requires mechanisms that can assist the system?s design and development. This paper presents a solution for assisting the design and development of CPS based on Model-Driven Development: MindCPS (doMaIN moDel for CPS) solution. MindCPS solution is based on a model that provides modelling primitives for explicitly specifying the autonomic behaviour of CPS and model transformations for automatically generating part of the CPS code. In addition to the automatic code generation, the MindCPS solution offers the possibility of rapidly configuring and developing the core behaviour of a CPS, even for nonsoftware engineers. The MindCPS solution has been put into practice to deploy a smart metering system in a demonstrator located at the Technical University of Madrid.