Multi-Objective Design Space Exploration for Cyber-Physical Systems satisfying hard Real-Time and Reliability Constraints

Weber, Raphael and Henkler, Stefan and Rettberg, Achim
Proceedings of IDEAL'14 Workshop
Developing embedded safety-critical systems was and remains a challenging task. Embedded systems engineers are supposed to design solutions that satisfy company shareholders, customers, certification authorities, etc. With these many partly contradicting demands it is very hard to find an optimal or optimized solution. Under special circumstances it sometimes is even impossible to find a viable embedded system design. In general, Pareto optimization is a well-known method to solve multi-objective optimization problems. However, when applying this method the user cannot influence this notion of optimality to address his own needs and preferences. This means that different objectives cannot be constrained or prioritized with respect to their optimization targets. In this work, we therefore propose an iterative design space exploration approach. We present an extension for our previously published semi-automated design space exploration (DSE) for safety-critical embedded real-time systems [1]. The previous approach is able to consider mapping constraints as well as the objective of minimizing costs. We extend this approach to a multitude of constraints and objectives by supporting an iterative refinement of the objectives as well as the constraints for the design space.
CPSWeek 2014 - IDEAL'14 Workshop
IFIP Springer Series
Software Platform Embedded Systems 2020_XT