Vibration Suppression And Contour Tracking Control Of Scanning Motions For Nanopostioners

A piezo-actuated nanopositioner with nanoscale resolution is one of the key components for scanning probe microscope(SPM).High-speed and high-precision are the basic requirements for scanning motions of piezo nanopositoners.The flexible hingle mechanism is always adopted in the design of naopositioners,which induces the lightly damped modes into the system and may lead to undesired vibrations.Thus,the working bandwidth of piezo-actuated nanopositioners is often limited between about 1%～10%of its first mechanical resonance frequency because of the lightly damped modes,which degrades the performance of high-speed.On the other hand,contour tracking error rather than individual axis tracking error is an important index for motion precision in the process of scanning motions.For multi-axis precision motions,traditional approaches focus on individual axis control.However,a poor synchronization between relevant axes will reduce the controur tracking accuracy.In another word,good individual tracking performance can not guarantee good contour tracking performance.The existence of contour error degrades the high-precision performance of piezo-actuated nanopositioners.For this,this dissertation is to address the typical scanning motions for piezo nanopositioners through vibration rejection and contour tracking control.To improve the performance of high-speed and high-precision simultaneously,an integrated feedforward contour tracking controller with a feedback vibration controller is studied in this dissertation.Firstly,four types of resonant controllers based on Negative-Imaginary(NI)theory are studied by comparasions.Some parameters are hard to be tuned or dertermined in these controllers.For this,a hybrid model-driven and data-driven method is proposed for the tunning of parameters in the positive velocity and position feedback(PVPF)controller,which relatively has more parameters among the aforementioned NI controllers.On the basis of that,a set of transfer models under variant loads are identified through step response method.A robust resonant controller(RRC)is then proposed to improve the robustness versus load variance.The parameters in the RRC are determined through an analytical approach.The controller gains are selected based on the small gain theory.A set of experiments compared with traditional IRC are conducted to demonstrate the effectiveness of the proposed RRC.Results show that the closed-loop bandwidth is improved by 27%from traditional IRC to the proposed RRC control.Moreover,the proposed RRC performs better under the uncertainties caused by load variation from 0～1000 g.Secondly,inspired by position domain control concept,a synchronized iterative learning control(SILC)is proposed to improve the contour tracking accuracy.The monotonic convergence condition is checked via computation of the lifting matrics of the control system,a set of parameters are then determined.Both simulations and experiments are performed to verify the effectiveness of the proposed SILC by comparations with traditional ILC.Results demonstrate that the proposed SILC achieves higher tracking precision compared with traditional ILC in time domain with an improvement of 49%,61%and 65%for sector,parabola and spiral contours respectively.Lastly,a two-input-two-output(TITO)model of the system is identified,where the lightly damped modes exist in the high frequency regions of the cross coupled parts.A static decouping matrix is designed to relieve the coupled lighly dmaped modes.After this,a TITO controller integrating feedforward SILC and feedback RRC is designed.Simulations are conducted to verify the effectiveness of the parameters determined by the monotonic convergence condition.A set of tracking experiments for raster,spiral,cycloid and Lissajous scanning signals at 20,40 and 60 Hz are carried out to evaluate the performance of the designed TITO controller.Results show that the integrated feedforward contour tracking control and feedback damping control scheme bring a significant improvement in tracking errors with comparision to feedback damping control alone.Tracking errors for the four typical scanning signals are all reduced by more than 75%.The new methods for damping controller parameters’ tuning and new algorithms for rapid and fast tracking control for piezo nanopositioners are investigated in this dissertation.Based on these efforts,scanning motions in one-axis and two-axes of the piezo nanopositioners are studied and analyzed through in-depth experiments.On one hand,the contributions of this dissertation will have application prospects in the field of high-speed high-precision motion controllers’ R&D for smart materials driven nanopositioners.On the other hand,the efforts can be introduced into the area of micro and nano-fabrication,multi-axis cooperative positioning,manipulation and alignment control in the field of information,material,biology and medical treatment,etc.