Analysis And Design Of Global Adaptive Tracking Control For Uncertain Nonlinear Systems

This dissertation investigates the problem of global adaptive practical out-put tracking for several classes of uncertain nonlinear systems. Mainly by the approaches of adding a power integrator and backstepping, and the ideas of uni-versal control and dead zone, global adaptive state-feedback tracking controllers and global adaptive output-feedback tracking controllers are designed respectively for two classes of high-order nonlinear systems with unknown control directions, and for two classes of uncertain nonlinear systems with unknown growth rate. The main contents of this dissertation are composed of the following four parts:(I) Global practical tracking control design for high-order uncertain nonlinear systems with unknown control directionsThis part addresses the global practical tracking problem for a class of high-order uncertain nonlinear systems with unknown control directions under fairly weak conditions. Motivated by the recent tracking works and the ideas of universal control and dead zone, by the approaches of Nussbaum-gain and adding a power integrator, and introducing a novel coordinate transformation, new updating laws, and pseudosign functions, the basic assumption that the control directions are required to be known in closely related tracking works is successfully removed, and a continuous adaptive state-feedback tracking controller is proposed. The designed controller is completely explicit, and guarantees that the tracking error becomes prescribed arbitrarily small after a finite time, and simultaneously all the closed-loop states are bounded.(II) Global practical tracking control design for a large class of high-order uncertain nonlinear systemsThis part addresses the global practical tracking problem for a more gen-eral class of high-order uncertain nonlinear systems, and particularly extends the unknown control coefficients with lower and upper constant bounds in the first part to the ones which may take arbitrarily large and/or small values. By the approaches of Nussbaum-gain and adding a power integrator, and introducing new updating laws, and a continuously differentiable function with a relatively simple structure, the new control design scheme of global adaptive state-feedback practical tracking is proposed. It is shown that the designed controller guarantees that all the closed-loop system states are bounded and the tracking error becomes prescribed arbitrarily small after a finite time.(â…¢) Global practical tracking by output-feedback for nonlinear sys-tems with unknown growth rateThis part is devoted to the global practical tracking by output-feedback for a class of uncertain nonlinear systems which have unmeasured states dependent growth with unknown constant rate. Motivated by the related stabilization re-sults, flexibly using the ideas of universal control and dead zone, and introducing an observer with a new dynamic high-gain, a global adaptive output-feedback tracking controller is designed. Moreover, with the aid of the proposed new pat-tern of performance analysis, the designed controller can make the tracking error prescribed arbitrarily small after a finite time while keeping all the states of the closed-loop system bounded.(â…£) Global practical tracking via output-feedback for cascade sys-tems with unknown growth rate and unknown control coefficientsThis part considers the global practical tracking problem by output-feedback for a class of uncertain cascade systems with more uncertainties. The systems in-vestigated are substantially different from the closely related work. The problem becomes much more difficult to solve due to the simultaneous existence of zero-dynamics, unmeasured states dependent growth with unknown constant rate, and unknown time-varying control coefficients with unknown constant bounds. By skillfully combining the methods of universal control and dead zone with backstep-ping technique, and introducing a distinct observer with a new dynamic high-gain, a new adaptive tracking control design scheme is successfully proposed. With the help of the pattern of performance analysis proposed in the third part, it is shown that the designed controller guarantees that all states of the resulting closed-loop system are globally bounded and the tracking error remains prescribed arbitrarily small after a finite time.The above four parts also give the corresponding simulation examples, re-spectively, to illustrate the correctness and effectiveness of the proposed control design schemes of global adaptive practical tracking.