Complex control systems are heterogeneous, because of discrete computer-based controllers interacting with continuous physical plants, regular data sampling interleaving with usually irregular interaction. This heterogeneity imposes great challenges for control system design technologies because of end-to-end control performance modeling and simulation from algorithms to software/hardware implementation, maybe lose the parallelism between the model and the code, lose the reuse of the software component, increase the designing complexity of the software, which is also weak, and it's extremely hard to implement in different platforms.In this paper, I have compared model-based design and platform-based design, have analyzed the benefit of the model-based design, have introduced the concept of actor and computer model, and have raised the actor-oriented design method to solve these problems by separating the data-centric computational components and the control-flow-centric scheduling and activation mechanisms. I have introduced some important computing models which use formal models of computation to manage the interactions among actors, different frameworks can be composed hierarchically to manage heterogeneity, improve understandability, and achieve actor and framework reusability.In this paper, I have researched an actor-oriented software design methodology to use platforms with useful modeling properties to specify designs, to orthogonalize the actor definition language and the actor composition language, enabling highly polymorphic actor definitions and design using multiple models of computation. This methodology is implemented through the Ptolemy II software environment. As an example, the methodology and the Ptolemy II software have been applied to the design of a single pendulum simulating system with period running. Separating platform-independent functionality and timing issues from platform-dependent scheduling andcommunication issue and time-triggered mode switches, whose predictability makes it particularly suitable for heterogeneous embedded control systems with hard real-time constraints. |