Pneumatic Drive System Identification And Adaptive Vibration Control Of Flexible Manipulator

Industrial robot manipulators are general-purpose machines used for industrialautomation in order to increase flexibility, productivity, and product quality. Among the rigidand flexible manipulator types, attention is focused more towards flexible manipulators.Compared to the rigid manipulator, flexible manipulators have the potential advantage oflower cost, higher operational speed, greater payload-to-manipulator-weight ratio, smalleractuators, lower energy consumption, better maneuverability, better transportability and saferoperation. The greatest disadvantage of flexible manipulators is the vibration problem due tolow stiffness and low damping. The large magnitude vibration of flexible manipulator will lastfor a long time. Active vibration control has been considered as a solution for this problem.Active vibration control of pneumatic drive and PZT drive flexible manipulator isinvestigated in this research.Firstly, the experimental setup is described. Depending on the control valve used inpneumatic drive, a PCM valve based pneumatic drive flexible manipulator vibration controlsystem and a proportional valve based pneumatic drive flexible manipulator vibration controlsystem are described. A hybrid driving scheme is achieved by composing the proportionalvalve based pneumatic drive and piezoelectric actuator.Secondly, an adaptive phase adjusting controller is developed for the PCM valve basedpneumatic drive flexible manipulator system. The adaptive phase adjusting controller aims tocompensate the phase lag introduced by pneumatic drive. Large time delay exhibited bypneumatic driver, and a low damping system has a small phase margin. The closed-loopcontrol system will become unstable if the phase of the controller is not tuned appropriately.The adaptive phase adjusting controller can optimize its phase, according to a controlperformance index online, and compensate the phase lag introduced by pneumatic driveadaptively.And then, the system dynamics models of the hybrid drive system are obtained based onsystem identification approaches. Nonparametric estimation and parametric estimationmethods are used in system identification of the model of piezoelectric manipulatorsubsystem. Due to the high nonlinearity of the pneumatic drive system, a single global linearmodel is not accurate to describe its dynamics. A self organizing map based multi-modelsystem identification approach is used to get multiple local linear model of the pneumaticdrive system.Finally, a self organizing map based multi-model inverse dynamic controller and a variable damping pole-placement controller are investigated in accordance with the differentcharacteristics of pneumatic drive and piezoelectric actuator. The models used in controllersdesign are got by system identification. Experiments of pneumatic drive flexible vibrationcontrol, piezoelectric manipulator vibration control and hybrid vibration control areconducted. The experimental results are compared with the results of PD controller. The selforganizing map based multi-model inverse dynamic controller and the variable dampingpole-placement controller can improve the performance of flexible manipulator vibrationcontrol, as compared with the corresponding investigating.