Unifying Model-based and Reactive Programming within a Model-based Executive

Real-time model-based eduction has recently emerged as a vital component in Al’s tool box for developing highly au-tonomous reactive systems. Yet one of the current hurdles towards developing model-based reactive systems is the num-ber of methods imultaneously employed, and their corre-sponding melange of programming and modeling languages. This paper offers an important step towards unification of re-active and model-based programming, providing the capabil-ity to monitor mixed hardware/software systems. We intro-duce RMPL, a rich modeling language that combines prob-abilistic, constraint-based modeling with reactive program-ming constructs, while offering a simple semantics in terms of hidden state Markov processes. We introduce probabilis-tic, hierarchical constraint automata, which allow Markov processes to be expressed in a compact representation that preserves the modularity of RMPL programs. Finally, a model-based executive, called RBurton is described that ex-ploits this compact encoding to perform efficent simulation, belief state update and control sequence generation.