In this project we will study theories and methods for probabilistic modelling, analysis, and prediction of control-system software executing in resource-constrained embedded systems.
We will adopt a component-based approach to control-system development. Using components as the fundamental unit of reuse, we will employ run-time monitoring techniques to extract probabilistic models of the component behaviour. Combining the behavior models and the architectural model of a component assembly we will derive stochastic properties, such as reliability, expected delays, and resource consumption. We will extend and combine methods from disparate disciplines, such as probabilistic reliability predictions, stochastic scheduling analysis and software component technologies. It is our thesis that software components are a suitable vehicle to introduce advanced analysis techniques in a software-engineering context for embedded control-systems; a feat that has yet to be accomplished.
Contract-Based Reusable Worst-Case Execution Time Estimate (Aug 2007) Johan Fredriksson, Thomas Nolte, Mikael Sjödin, Heinz Schmidt Proceedings of the 13th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA07)
Reusing Worst-Case Execution Time Analysis with Component Contracts (Aug 2007) Johan Fredriksson, Thomas Nolte, Mikael Sjödin, Heinz Schmidt Proceedings of the 9th Real-Time in Sweden (RTiS07)
Clustering Worst-Case Execution Times for Software Components (Jul 2007) Johan Fredriksson, Thomas Nolte, Andreas Ermedahl, Mikael Sjödin Proceedings of the 7th International Workshop on Worst Case Execution Time Analysis (WCET07)
Contract-Based Reusable Analysis for Software Components with Extra-Functional Properties (Jul 2007) Johan Fredriksson, Thomas Nolte, Andreas Ermedahl, Mikael Sjödin Proceedings of the Work-In-Progress (WIP) session of the 19th Euromicro Conference on Real-Time Systems (ECRTS07)