Limited data rate in control systems with networks (Q698690)

This book, based on the first author's Ph.D. thesis, has been motivated by the recent progress in communication technology and wide use of networks. It deals with the issues arising in control design for systems involving communication networks. The typical problem studied can be stated as follows: given a plant, a communication channel with limited data rate, and control objectives, find a controller that uses the channel in the feedback loop such that the system achieves the control objectives. In Chapter 2, which follows a brief introductory Chapter 1, several basic facts related to control networks are presented. Traffic in such a control system is characterized, and differences between the requirements for data networks and control networks are discussed. The advantages for using the control networks are also pointed out. Practical issues in control networks are summarized, and some ideas on the current technology and networking are provided. Some industrial examples are given. Chapter 3 is concerned with a distributed control problem, where the plant is controlled by two local controllers which can communicate to each other over a limited data rate channel. The methods for time sequencing of messages over the channel are studied and stability issues are characterized. The advantages of the use of networks in the context of decentralized control are indicated. Chapter 4 deals with the design of a stabilizing controller for a single-input linear time-invariant plant that has a finite data rate channel in the feedback loop. First, a design method for a sampled-data controller that stabilizes the plant in a quadratic manner is proposed. Then the result is extended to a design problem of a quantizer in a sampled-data controller. A design example of a magnetic levitation ball system is discussed, and the data rates necessary for stabilizing using several quantizers are compared. In Chapter 5, the results of the previous chapter are extended to obtain methods for reducing the data rate necessary for stabilization of a plant. This can be achieved either by decreasing the number of bits or the number of partition cells, or by increasing the time between samplings. The methods for time-varying quantization and nonuniform sampling are suggested, while combinations of these techniques, which can be rather complicated, are not discussed. The goal of Chapter 6 is to extend the problems studied in Chapters 4 and 5 to the case of multiple inputs considering first sampled-data control, then the quantized sampled-data, and finally the dwell-time switching control. Some directions for future research are listed in the last chapter, and the bibliography contains 99 entries.