講演会開催報告: The model-based perspective of dependable systems design
Can control engineers do dependability?
The model-based perspective of dependable systems design
Klaus Janschek, Professor
Institute of Automation, Technische Universität Dresden, Germany
Robotic systems show inherent complexity, both in terms of functionality and integrated heterogeneous technologies. This makes robotic systems highly critical subject to failures at different technological levels (software, hardware, human operator). An important requirement to the systems design is high dependability of such systems, i.e. ensuring most reliable and safe operation under presence of non-avoidable threats.
This talk discusses modern model-based design aspects for ensuring appropriate dependability properties already at the level of control systems design, i.e. giving the control engineers transparent means for analyzing control systems architectures and proactively designing a dependable control systems solutions.
Current research results of our TU Dresden Automation Engineering Lab demonstrate the successful applicability of model-based system threat analysis to control systems for robotic systems introducing new concepts such as probabilistic error propagation using dual graph error propagation model in combination with Markov chain models and probabilistic model checking techniques, model-based redundancy using fault tolerant design patterns and model-level assessment of dependability metrics.
The developed methods have been implemented in a powerful computer-aided framework allowing automated analysis of UML/SySML or Matlab/Simulink based system models.
6月26日（水） 9：15 – 10：15