Search

It is well known that a stable discrete-time control system may become unstable when the digital controller is actually implemented with a digital control processor (DCP) due to the finite word length (FWL) effects and that the stability behavior of the system depends on the structure of the digital controller. Due to the difficulties in adopting the existing stability robustness measures, a new measure is derived, which is tractable and hence enables us to handle the digital controller structure problem easily. The optimal controller structure problem is formulated by maximizing this measure. An analytical solution is presented. An algorithm is developed for finding sparser structures. Index Terms - Optimal structure, pole sensitivity, stability.

The discontinuous perturbation analysis (DPA) method is presented to deal with discontinuous sample performance functions. This method is applicable to derivative estimation with respect to most threshold-type problems in discrete-event dynamic systems (DEDS). Through modeling of discontinuities by step function, we are able to provide a unified framework for constructing the derivative estimation of a DEDS. As a result, we offer an alternate approach to derive derivative estimation in DEDS's. Finally, some relationships with other PA techniques are discussed, and a numerical example is presented.

In this paper we study sufficient observation and work spaces for supervisors of a class of discrete-event dynamical systems (DEDS). We use finite state automata to model a DEDS. The finite automata generates a formal language defined over the set of events in the DEDS. A supervisor is a feedback system that observes a generated trace of events and dynamically disables (or enables) a subset of controllable events such that the closed-loop system behaves as desired. We model the desired behavior of the DEDS by a sublanguage defined over the set of events. A sufficient observation space is a collection of events whose observation by a supervisor is enough to realize a given desired behavior. A sufficient work space is a sufficient observation space such that the control action of the supervisor is limited to only the controllable elements of the work space. In many cases, there exists a sufficient work (or observation) space that is smaller than the set of all events. In this work, we construct algorithms to evaluate a sufficient work (or observation) space. The reduction of work (or observation) spaces results in reducing the number of event detectors, communication, and command channels between the supervisor and the plant.

Within this brief paper, a stable indirect adaptive controller is presented for a class of interconnected nonlinear systems. The feedback and adaptation mechanisms for each subsystem depend only upon local measurements to provide asymptotic tracking of a reference trajectory. In addition, each subsystem is able to adaptively compensate for disturbances and interconnections with unknown bounds. The adaptive scheme is illustrated through the longitudinal control of a string of vehicles within an automated highway system (AHS).

We derive the optimal LQR controller for asynchronous multirate digital control. Independent processors operating without synchronization update different components of the control vector at Poisson arrival times. An iterative algorithm to compute the steady-state control law and optimal cost is described and its convergence is demonstrated.

The following problem is addressed: Given a continuous-time plant, with continuous-time performance objectives, expressed in terms of the L(super Infinity)-induced norm, design a digital controller that delivers or optimizes this performance. This problem differs from the standard discrete-time methods in that it takes into consideration the inter-sample behavior of the closed-loop system. The resulting closed-loop system dynamics consist of both continuous-time and discrete-time dynamics and thus such systems are known as hybrid systems. It is shown that given any degree of accuracy, there exists a standard discrete-time l(super 1) problem, which can be determined a priori, whose solution yields a controller that is almost optimal in terms of the hybrid L(super Infinity)-induced norm. This is accomplished by first converting the hybrid system into an equivalent infinite-dimensional discrete-time system using the lifting technique in continuous time, then the infinite-dimensional parts of the system which model the inter-sample dynamics are approximated. We present a thorough analysis of the approximation procedure, and show that it is convergent at the rate of (1/n). Explicit bounds that are independent of the controller are obtained to characterize the approximation. Finally, it is shown that the geometry of the induced norm for the sampled-data problem is different than that of the standard l(super 1) norm, and hence there might not exist a linear isometry that maps the sampled-data problem exactly to a standard discrete-time problem.

This paper investigates the properties of the mapping from the simple zero [Gamma] of a scalar continuous-time system to the corresponding zero [Gamma](T) of the sampled-data system that results by its discretization using a zero-order hold, where T is the sampling period. It is shown that [Gamma](T) admits a Taylor expansion with respect to T, and that it coincides with that of exp([Gamma]T) at least up to the second-order term, in general, and at least up to the third-order term if the relative degree of the continuous-time system is greater than or equal to two. The result is applied to derive a new stability condition of [Gamma](T) for sufficiently small sampling periods.

A quantitative stability condition for a sampled-data feedback system is formulated. The system is obtained from a stable continuous-time feedback system by digitalization of the controller. The condition shows how to determine a sampling frequency [f.sub.0] such that for all f larger than [f.sub.0] the discrete-time system remains stable.