This paper presents a dynamic output feedback control with adaptive rotor-imbalance compensation based on an analytical Takagi-Sugeno fuzzy model for complex nonlinear magnetic bearing systems with rotor eccentricity. The rotor mass-imbalance effect is considered with a linear in the parameter approximator. Through the robust analysis for disturbance rejection, the control law can be synthesized in terms of linear matrix inequalities. Based on the suggested fuzzy output feedback design, the controller may be much easier to implement than conventional nonlinear controllers. Simulation validations show that the proposed robust fuzzy control law can suppress the rotor imbalance-induced vibration and has excellent capability for high-speed tracking and levitation control.

Applications of neural networks to the identification and control of flexible manipulators are considered. Usually, neural networks need a large number of neurons and a great amount of computation for learning, and the error is not easy to reduce. This study tries to combine the a priori knowledge of the corresponding rigid manipulator's model with two multilayered neural networks for the identification and control of a flexible-link manipulator. The suggested approach can use fewer neurons and needs shorter learning time for reducing the error. A planar (in the vertical plane) two-link flexible arm with the first link rigid and the second link flexible is tested via simulation. The mathematical model of the flexible arm for simulation is derived by the finite element method using Lagrange's equation.

A two-stage control strategy, composed of rigid-body-motion-based partial feedback linearization and state feedback H/sub infinity / control for robotic manipulators with flexible joints and links is presented. The partial feedback linearization control synthesized by the differential geometric structure algorithm is an input-output decoupling and linearization feedback law capable of precise rigid-body-motion-based trajectory tracking, but the zero dynamics of the unobservable nonlinear elastic subsystem remains unstable. In order to actively suppress the elastic vibrations, it is necessary to sacrifice the perfect rigid-body-motion-based tracking capability somewhat by introducing a robust stabilization control in the vicinity of the desired trajectory. The use of state feedback H/sub infinity / control with integral action for the synthesis of the robust stabilization control is able to achieve active damping of elastic vibration and robust tracking of motor dynamics. A two-link arm is used in the numerical study.>

A motor-based decoupling and partial (input-output) linearization approach to the control of multilink robots with joint and link flexibilities is studied. The control strategy consists of two parts; nominal tracking control and perturbed stabilization control. The nominal tracking control derived by the differential geometric structure algorithm is an input-output decoupling and partial linearization feedback law capable of precise motor-based trajectory tracking, but the zero dynamics of the unobservable nonlinear elastic subsystem remains unstable. In order to actively suppress the elastic vibrations, a perturbation control is introduced in the vicinity of a desired trajectory. The perturbed stabilization control synthesized by the combined LQR (linear quadratic regulator) and servocompensator is used to achieve active damping of elastic vibration and robust tracking of motor dynamics. To offset the tracking errors of the end effector caused by joint and link deflections due to gravity, the quasi-static deflections can be taken into account in the trajectory planning and LQR. A two-link arm is tested by simulation.>