Scenario-based Synthesis of Distributed Mechatronic Systems

When developing safety-critical mechatronic systems, a faulty implementation can lead to high costs. Typically the costs caused by faults increase the more the later they are found during the development process. As faults can already be caused by inconsistent requirements towards the developed system, these should be checked for inconsistencies as soon as possible. For being able to do this automatically using software, formalizing the requirements is essential.

A formalization of this kind is part of the scenario-based development process for mechatronic systems developed at this research group. Within this process first a formal specification consisting of communication scenarios is developed. Subsequently, an automatic "synthesis"-approach either finds a contradiction within the requirements or generates a basic implementation of this specification as a proof for its realizability. However, this implementation realizes a global controller for the system as a single automaton model. This kind of central control can not directly be applied within a distributed system.

The goal of this PhD project is to realize a "distributed synthesis", an approach for generating a distributed implementation from a global specification. This distributed implementation describes the communication behavior of the individual cooperating subsystems each by its own automata model. Furthermore, this generated model is required to be compatibel to our MechatronicUML modeling approach. This way it can be used as an initial basis for further manual development using the MechatronicUML approach.

Contact: Christian Brenner