Adaptive Control And Sampled-data Control For Nonlinear Systems With Input Quantization

In practical application,most controlled systems are nonlinear,and most modern con-trollers are digital or networked to the systems.Hence,the adaptive control for the nonlinear systems with input quantization and the sampled-data control for the nonlinear systems are very important from the point of control theory and practical application.On the basis of the existing work,based on fuzzy logic systems and neural networks,this thesis investigates two types of control problem for the nonlinear systems:one is the adaptive control for the nonlinear systems with input quantization;another one is the sampled-data output feedback control for the nonlinear systems.The main contents are summarized as follows:1.The adaptive state feedback control problems are investigated for high-order nonlin-ear systems with unknown control directions and input quantization.For the high-order upper-trigger nonlinear system with unknown control directions,we introduce the Nussbaum-type function together with backstepping design method,design an adaptive controller.For the nonlinear system with input quantization,we design an adaptive con-troller based on hysteretic quantizer and homogeneous domination.The designed con-trol methods ensure that all the signals in the closed-loop system are bounded,and the states asymptoticly converge to zero.The simulation examples are provided to demon-strate the applicability of the proposed design schemes.2.The problem of finite-time adaptive control of a class of strict-feedback nonlinearly parameterized systems with quantized input signal is studied.A hysteretic type of quan-tizer is used to avoid chattering.By using adding a power integrator technique,a new controller is designed to ensure the global finite-time stability of the nonlinear system.Finally,the simulation examples are given to demonstrate the effectiveness of our design method.3.The fuzzy adaptive control and neural network control for the nonlinear systems with input quantization are considered.For a class of uncertain nonlinear systems with a quantized signal,fuzzy logic systems are used to approximate nonlinear terms without imposing prior matching conditions required.A new adaptive backstepping controller is designed to guarantee that the underlying uncertain nonlinear system is semiglobally uniformly ultimately bounded.In addressing the adaptive neural backstepping control for multiple-input and multiple-output nonlinear systems in pure-feedback form with in-put quantization,we construct a high-gain state observer and an output-feedback adap-tive control scheme using backstepping method,with neural networks to estimate the un-certain nonlinear functions.Then,we propose an output feedback neural controller that ensures all the state trajectories in the time-delay quantized nonlinear systems are ulti-mately bounded,with the control signal being quantized by either a hysteretic quantized or a logarithmic quantized.Illustrative examples are presented to show the applicability of the new control method developed.4.The problem of the sampled-data control for the nonlinear systems with disturbance is investigated.For the sampled-data fuzzy control of nonlinear systems in strict feedback form with disturbances and random missing input data,we propose a novel method in which a state observer and a disturbance observer are combined to construct a sampled-data fuzzy output feedback controller.The stochastic variables with a Bernoulli dis-tributed sequence are used to model missing input data.Fuzzy logic systems are applied to approximate nonlinearities without requiring prior knowledge.The relation between observer gain and sampling period is established.The proposed sampled-data controller can guarantee that the nonlinear systems are stable.Example is given to show the effec-tiveness of the proposed new design method.