Showing papers in "Control and dynamic systems in 1985"

Decentralized Control Using Time-Varying Feedback

Abstract: I. FIXED MODES AND DECENTRALIZED CONTROL . . . A. WHAT ARE FIXED MODES?.. . . . . . . . B. DECENTRALIZED FIXED MODES AN EXAMPLE c. CHARACTERIZATION OF FIXED MODES . . . . II. ELIMINATING DECENTRALIZED FIXED MODES WITH TIME-VARYING CONTROLLERS . . . . . . . A. THE TECHNICAL LEMMAS . . . . . . . . . B. FORMAL RESULTS . . . . . . . . . . . . c. MULTICHANNEL SYSTEMS . . . . . . . . D. AREAS FOR FURTHER STUD; . . . . . . . . REFERENCES . . . . . . . . . . . . . . . . . . . . . 85 . . . . . 85 . . . . . 89 . . . . . 90

Distributed estimation for large-scale event-driven systems

Abstract: This paper is concerned with distributed estimation theory for dynamic stochastic systems. To bring into mathematical focus the issues arising from the absence of centralized information, we shall consider the class of event-driven stochastic dynamic systems. These systems are hybrid state systems whose state consists of a discrete component and a continuous component, where the transitions between the discrete components of the state correspond to the occurrence of the events. The evolution of the events in time is modeled by a Markov chain; the states of the Markov chain influence the dynamics of a linear stochastic system which models the evolution of the continuous components of the state. The distributed estimation problem is to estimate the hybrid state using noise corrupted, distributed observations. 22 refs., 6 figs.