The Model Based Design approach for Hybrid Systems control software synthesis is particularly appealing since Formal System Level Specifications are usually much easier to define than the control software itself. In this setting, Design Space Exploration has the goal to find a suitable (with respect to costs and performance) choice for system design parameters. Unfortunately, a substantial part of the time devoted to design space exploration is spent trying to solve control software synthesis problems that do not have a solution. We present an on-the-fly algorithm to control software synthesis that enables effective design space exploration by speeding-up termination when no controller is found. Our experimental results show the effectiveness of our approach and how it can support a concrete realizability and schedulability analysis.
展开▼
