Extension of Finite State Machine


Researchers: Rachel Zhou
Advisor:Edward A. Lee

The current Ptolemy II software uses modal model to compose finite state machine (FSM) with other models of computation. A modal model is a composite actor with an FSM director as its local director. A modal model can be embedded in, as well as each of its states can be refined to a different domain. The FSM itself cannot represent the computational dynamics or the consumption and production of tokens. But rather they are inferred by the current state refinement. Such semantics limit the designer's power to specify the behavior of FSM/modal model.

In this project, we would like to experiment adding expressiveness in state transitions of FSM. Transitions should be able to infer the consumption and production of tokens. This requires correctly parsing the guard and action expressions in the transition. The study plan is as follows: (1) We will first study the case where all the inferred ports have rate 1. Ports that are not inferred have rate zero. (2) Next we shall design a syntax where multirate ports can be specified. We assume there is no state refinement in the above two cases. (3) We study the case where both transition and state refinement consume and/or produce tokens.

References:

[1] C. Brooks, E. A. Lee, X. Liu, S. Neuendorffer, Y. Zhao, H. Zheng "Heterogeneous Concurrent Modeling and Design in Java (Volume 3: Ptolemy II Domains) ," Technical Memorandum UCB/ERL M04/17, June 24, 2004, University of California, Berkeley, CA USA 94720.

Last updated 11/01/04