On Optimal Performance Of Networked Control Systems With Communication Constraints

Networked control systems (NCSs) have been found successful applications in a wide range of areas such as industrial automation, distributed/mobile communication, and unmanned vehicles. While NCSs have received increasing research attention, they have also given rise to new challenges due to inherent communication constraints. Such as bandwidth constraints, signal-to-noise ratio, packet dropouts and quantization precision, time delays, will degrade the performance of NCSs or even cause instability. Now, the most research interesting focuses on modeling of the NCSs and stabilization analysis, however, the optimal tracking performance and optimization design have less been considered with the constraints of communication channel parameters, and the optimal tracking performance in control design has been an important area of research. Currently, most results have pointed out that the input signal and an inherent characteristic of system decision performance limits of control system. The tracking performance problems of networked control system are studied in this dissertation based on the communication constraints. The main contents of this dissertation are outlined as follows.The control systems with measurement noise and disturbance rejection are considered in this dissertation, which investigates the optimal performance of tracking a stochastic signal for linear time-invariant systems. The display expression of optimal tracking performance is obtained by applying the two-freedom-degree controller and spectral factorization method. It is shown that the non-minimum phase zeros, statistical characteristics of the reference input signal, the unstable poles of a given plant and disturbance signals affect a feedback system's optimal tracking ability. The results also provide theoretical guidance for control systems design.The optimal tracking problem for multiple-input multiple-output linear time-invariant discrete-time systems with communication constraints are studied in this dissertation. The reference input under consideration is a step signal and the tracking error power between the system output and reference signal was adopted as performance index. An exact expression of the optimal tracking performance is obtained by applying the two-freedom-degree controller. It is shown that the non-minimum phase zeros, zero direction, unstable poles, pole direction of the given plant, and reference input signal direction affect the optimal tracking performance. The results also show how the optimal tracking performance is limited by the bandwidth and additive white Gaussian noise of communication channel.The network-induced delay can result in the performance degradation of control systems and even worse cause a system to become unstable. Therefore, the network-induced delay should be taken into consideration when it comes to the tracking problem. In this dissertation, we investigate the optimal tracking for single-input single-output linear time-invariant systems and discrete-time systems, with network-induced delay in communication channel. The exact solution of optimal tracking performance is obtained by using the single degree of freedom controller. It is shown that the optimal tracking performance is constrained by nonminimum phase zeros, unstable poles of a given plant and network-induced delays and that how the optimal tracking performance is limited by the network-induced delay.The communication capacity, which is determined by the signal-to-noise ratio (SNR), has a direct impact on the performance of the NCSs. The optimal tracking performance of single-input-single-output NCSs based on SNR constraints is proposed in this dissertation. The tracking performance is measured by the energy of the error variance response between the output of the plant and the reference signal. The display expression of optimal tracking performance is obtained by H2 norm and spectral factorization technique. It is shown that the optimal tracking performance is constrained by unstable poles of a given plant, the power spectral density of a given reference signal and SNR of communication channel. The result obtained in this work explicitly show how the optimal tracking performance is limited by the SNR of communication channel.The performance of network control system is determined by control system itself and communication network. Thus, the optimal tracking problem for multiple-input multiple-output (MIMO) linear time-invariant (LTI) continuous-time systems with communication constraints and a code scheme is studied. Furthermore, the joint design of coder and controller is also considered. With the two-freedom-degree control structure, explicit expressions for optimal tracking performance are obtained. It is shown that the optimal tracking performance closely dependents on the nonminimum phase zeros, zero direction, the unstable poles, pole direction of the given plant, and the bandwidth and additive white noise of a communication channel. The results also provide a theoretical guidance for networked control systems design.Most network signal transmission is bidirectional, and communication code also seriously affects the performance of control system, thus, the optimal tracking problem of networked control systems with White noise and coding in a forward channel and channel feedback is studied. The performance is measured by the energy of the tracking error. The display expression of performance limitation of networked system is obtained by applying the two-freedom-degree control and the spectral factorization technique. The proposed results indicate that optimal tracking performance is only determined by the inherent characteristics of the given plant and communication parameters. The results also provide a theoretical guidance for networked control systems design. It is shown that the nonminimum phase zeros, the unstable poles of the given plant, and the code and white noise of a communication channel decides the optimal tracking performance of networked control systems.