Study On Robust Filtering And Control Method Of Nonlinear Systems

The actual control systems often exhibit some nonlinear characteristics, and saturation is typical and common. This presence of nonlinear characteristic saturation is not only likely to worsen the performance of systems but also may even lead to the instability of systems, and this characteristic is almost impossible to eliminate. On the other hand, a lot of practical control systems such as aerial devices, machine control systems robots and other engineering technologies satisfy the Lipschitz conditions. Additionally, there are external disturbances, changing in working conditions, various faults of systems and unmodeled errors in practical systems, so there exist uncertainty in the mathematical models of the objects. Therefore, further spread the research of Lipschitz nonlinear uncertain systems with saturation not only has theoretical significance, but also has great practical significance. With the development of computers and network technology, the study on the problems of robust filtering and control saturated nonlinear systems under the network environment is particularly important. Adding network-specific phenomena such as packet loss into saturated nonlinear systems undoubtedly increases the difficulty of research, and it becomes one of the issues to be resolved.Firstly, this thesis is for the nonlinear uncertain systems with time-varying delay and sensor saturation(output saturation), where the nonlinearity satisfies the Lipschitz condition, the saturation problem is solved by means of sector region method, and the uncertainty is set to be norm-bounded. By choosing the delay-dependent Lyapunov function, the 2L L?? performance criterion of the filtering error system is got, the appropriate robust 2L L?? filter is designed. Robust 2L L?? filter parameter matrices which guarantees the filtering error system to be asymptotically stable with an performance index are got by solving a convex optimization problem about 2L L?? noise attenuation level ?.Secondly, the robust ?H state feedback controller and dynamic output feedback controller of nonlinear uncertain systems with time-varying delay and actuator saturation are designed. Due to the sufficient condition of the presence of output feedback controller is non-strict linear matrix inequality, the matrix inequality solvability problem is cast into a convex optimization problem constrained by linear matrix inequalities according to the cone complementary linearization method. Thus, we can solve the parameters of controller using standard mathematical software.Then, the robust filtering control problems are studied for continuous-time nonlinear uncertain time-varying delay systems with saturation under network environment. The network-specific phenomenon——the packet loss is accounted, and the class of nonlinear systems are modeled as asynchronous dynamical system constrained by structural event rates. In the case of knowing packet loss rate, using asynchronous dynamical theory, robust filtering and control methods the 2l l?? performance criteria which guarantee the filtering error system exponential stability, and the H? performance criteria which guarantee the closed-loop asynchronous dynamic systems exponential stability are established. By solving convex optimization problems about noise attenuation level ?, the feasible filter, controller parameter matrices and the corresponding decay rates are got.Finally, based on the way of spacecrafts relative attitude with no relative measurement, using the robust H? filtering technique of optimal recursive filtering in the state linearization, the multiplicative error modified Rodrigues parameters are regarded as the state variables. And considering the presence of the delay from the network, spacecraft relative attitude algorithm design issue with the combination of gyroscope and Star Sensor determination is researched. Spacecrafts’ relative attitude errors are obtained by simulation, and the feasibility of the filter algorithm is verified.