Research On Modeling And Control Approaches Of Smart Material Actuators For Micro/Nano-manipulation

Nowadays,with development of technologies towards micro/nano scale,micro/nano-manipulation techniques have become more and more essential in both industrial and academic fields.Precise gripping technique is one of the most important technique in micro/nano-manipulation,it enhances precision and efficiency during object operation.Precise positioning capacity is another essential technique,which has been attractive for decades,it directly determines the micro/nano-manipulation precision.Recently,IPMC(Ionic Polymer Metal Composite)became one of the most ideal smart materials for micro-gripper fabrication,however,its instinctive nonlinearity and time dependent characteristics with respect to the output force and displacement hinder its implementation,and the control work is still challenging.For precise positioning,piezoceramic actuator has been always popular for its stable,fast merits and ultrahigh spatial resolution,however,its instinctive complex hysteretic characteristic brings much trouble to its positioning precision in micro/nano-manipulation process.This dissertation studies modeling and controller design approaches for regulating flexible IPMC actuator,and rigid piezoceramic actuator.For efficiently regulating IPMC actuators,which possess time-varying nonlinear characteristics,this dissertation takes a deep insight into these issues and proposes associated controllers to tackle them.For handling precision issues when using piezocramic actuator doing micro/nano-manipulation,this dissertation studies both feedforward contol method for open loop positioning and advanced feedback control approach for high freqeuncy periodic reference tracking operation.Flexible IPMCs and rigid piezoceramics are highly potential smart material actuators for conducting micro/nano-manipulation,this dissertation combines theoretical and experimental studies together,aiming at developing effective,precise control technologies to promote their development in this field.The main contents of this dissertation are briefly addressed as following:To carry out safe and effective manipulation on fragile micro-objects,this dissertation studies IPMC output force modeling and controlling problems.Back-relaxation(a generalized creep phenomenon)characteristic commenly exists in IPMCs,of which the generated force will change with time and the back-relaxation characteristic will also be influenced by the change of the water content or other environmental factors.To handle these issues,this paper presents a novel adaptive force control method(AIPOF,Adaptive Integral Periodic Output Feedback Control),based on IPMC generalized creep model of which parameters are obtained by using the FRLS on-line identification method.The AIPOF control method can achieve an arbitrary eigenvalue configuration as long as the plant is controllable and observable.This dissertation also designs a POF and IPOF controller as comparative studies.Simulation and experiments of micro-force-tracking tests are carried out,test results confirms that the proposed control method is potential for application of micro-gripper with force feedback.To generate displacement with high precision using flexible IPMC actuator,this dissertation builds up a hybrid model to describe IPMC instinctive back-relaxation(generalized creep phenomenon)property and hysteresis nonlinearity.To compensate the hysteresis and back-relaxation effects,this study first designs an adaptive inverse(AI)compensator considering the IPMC input-output relation varies with inside and outside circumstantial influences,such as time of usage,temperature,water content etc.Experimental study has been conducted to verify the AI compensation appraoch.To further reduce model uncertainty and environmental disturbaces,this dissertation combines a sliding-mode-like robust control module with the AI component,named discrete adaptive sliding-mode-like controller(DASMLC),to regulate the IPMC actuator.Stability of DASMLC method is analyzed,and comparative experimental study is conducted to show efficiency of this method and potential of implementation in micro-gripper for micro-manipulation.Piezoceramic actuators are usually employed in micro/nano-positioners for their high spatial resolution.However,instinctive complex hysteresis of piezoceramics significantly degrades their positioning precision.To effectively describe and further reduce the complex hysteresis phenomena,this dissertation develops a new type of modified Prandtl-Ishlinskii(PI)operator,named unparallel PI(UPI)operator,and the associated hysteresis model,EUPI model.To reduce complex hysteresis,an inverse compensation scheme is developed,and its stabilization method is proposed.To further enhance accuracy for implementations in nano/micro-manipulation systems,a compensation approach named extended scanning-range adapted hysteresis/creep hybrid(ESAH)method is proposed to handle different working ranges problem.Experiments on an Atomic Force Microscopy(AFM)based nano/micro-manipulation system are carried out to verify the approaches.Experimental results show that the ESAH method is viable for micro/nano-positioning tasks.To tackle positioning precision issue when using a piezoceramic actuator to track high frequency periodic reference in devices such as AFMs,this dissertation proposes an internal model,complex hysteresis cancellation method combined robust control scheme.First,this study models piezoceramic actuator precisely considing complex hysteresis and dynamics,and derives analytical expression of upper/lower amplitude boundary of the hysteresis component.Subsequently,the dissertation proposes discrete EUPI internal model(d-EUPI-IM)controller to handle periodic reference signal tracking problem,and the stabilization condition has been presented.To further tackle model uncertainty,a robust d-EUPI-IM controller design method is proposed,analysis shows tracking error of closed-loop system with robust d-EUPI-IM will converge with exponential speed.Comparative experimental study shows that this control method is effective and possesses potential for implementation of AFMs for fast scanning tasks.