Multivariable Adaptive Disturbance Rejection Methods And Aircraft Turbulence Compensation Techniques

In this dissertation,new adaptive control and disturbance rejection theory and techniques are developed,with their application to the aircraft turbulence compensation designs,for which two types of control problems are formulated from two different perspectives: one is for minimum phase systems,and one is for non-minimum phase systems.For minimum-phase systems with uncertain system parameters and unmatched disturbances,multivariable model reference adaptive control(MRAC)techniques are further developed,so that a new framework of MRAC based disturbance rejection theory is established,and is applied to solving the aircraft turbulence compensation problem.(i)Multivariable adaptive disturbance rejection schemes are developed using MRAC techniques for linear systems with a general interactor matrix,in the presence of uncertain system parameters and unmatched disturbances.We first derive key designed conditions in terms of the system interactor matrix,with which we develop nominal state feedback and output feedback based disturbance rejection controllers,based on the knowledge of system and disturbance parameters,which can ensure some ideal system stability and output tracking properties.Then,we develop a set of MRAC based disturbance rejection control techniques for systems with unknown parameters and unknown unmatched disturbances.To deal with the system high frequency gain matrix uncertainties,the LDU,SDU and LDS decomposition techniques of a matrix are used to relax the related design conditions.Such adaptive control designs can ensure that all signals are bounded and asymptotic output tracking is achievable in the presence of uncertain unmatched input disturbances and system parameters.(ii)An application to aircraft turbulence compensation is developed using a multivariable adaptive disturbance rejection algorithm.We conduct a complete study on nonlinear and linear models of aircraft dynamics under turbulence conditions,including a specific analysis of the turbulence effects.We define two sets of relative degrees for the disturbance to output subsystem and the control to output subsystem,and give related key design conditions for turbulence compensation.We derive a nominal state feedback based turbulence compensation controller for ideal output tracking and disturbance rejection.We develop an adaptive turbulence compensation scheme to ensure desired system performance in the presence of system and turbulence uncertainties.Simulation results are presented to show the effectiveness of our new method.(iii)An adaptive feedback linearization based disturbance rejection technique is developed for nonlinear systems with uncertain system parameters and unmatched disturbances.We specify key design conditions in terms of the control and disturbance relative degrees.We develop feedback linearization techniques to design a nominal disturbance rejection controller for ideal closed-loop system performance.We then design adaptive feedback linearization based disturbance rejection schemes for which we use an adaptive parameter projection algorithm to ensure the non-singularity of the control gain matrix estimate.We give a detailed system stability and tracking analysis,and demonstrate our algorithm for the turbulence compensation problem.For non-minimum phase systems with uncertain parameters from systems and unmatched input disturbances,we develop a set of adaptive control separation based LQ disturbance rejection theory and techniques and use them to solve the actuator failure compensation problem and aircraft turbulence compensation problem.(iv)An adaptive control separation based LQ disturbance rejection technique is developed for multivariable linear systems,using a modified Hamiltonian theory.Employing a new cost function,we establish a new control separation based LQ control framework for achieving both the feedback control and disturbance rejection goals.We systematically develop finite-time and infinite-time LQ control designs based on modified Hamiltonian functions,including a complete comparison analysis with traditional LQ results.We then develop an adaptive control separation based LQ disturbance rejection technique with stable adaptive laws,for system stability,disturbance rejection and output regulation,which is applied to solving the turbulence compensation problem for non-minimum phase aircraft system models.(v)An adaptive control separation based LQ actuator failure compensation scheme is developed for non-minimum phase discrete-time systems with uncertain dynamics and actuator failure parameters,to ensure desired system stability and output regulation.A new disturbance rejection problem is formulated for solving the actuator failure compensation problem through considering uncertain actuator failures as unmatched input disturbances.We establish a new control separation based LQ disturbance rejection framework by employing a new cost function,with which we develop an effective failure compensation design for generating a two-component control input: one component is for system stability and output regulation,and one component is for failure compensation.We systematically develop finite-time and infinite-time LQ control and state estimation based LQ control techniques,including their specific performance analysis.We develop an adaptive control separation based LQ failure compensation scheme for systems with uncertain parameters from systems and actuator failures,and give a complete stability analysis to demonstrate the effectiveness of actuator failure compensation.