Handling the design complexity of Cyber-Physical Systems

The objective of this thesis is to provide instruments useful to tackle the development of Cyber-Physical Systems (CPSs).Particularly attention will be given to the interaction between control algorithm and their implementation, that is, "Control/Scheduling Co-Design" and to the development of a testing tool for such systems. One of the research topics addressed in this work is the problem of how the scheduling of the control task could affect the performance of the whole system. The real-time research branch deals with scheduling and implementation problems, whilst the control research branch deals with control algorithms and their performance. These two worlds are now getting closer as the concept of "Cyber-Physical" systems pervades in the common engineering knowledge. Starting with linear systems, a modeling approach has been carried out to blend the two aspects and a performance index has been proposed to give insight on the system functioning. Another part of the thesis deals with the design of a simulation environment for autonomous vehicles. The high complexity of CPSs makes difficult to tackle everything by theoretical analysis and the deployment on a simulated plant is necessary to get confidence about the proper functioning of the system. The design of the simulation environment addressed the lack of complete suites for system simulation, integrating different aspects, such as multi-agent scenarios, a realistic virtual environment and hardware-in-the-loop capabilities.