Research On Precision Position/contact Force Tracking Control Methods For Piezoelectric-actuated Systems

With the rapid development of advanced manufacturing technology,biomedical engineering,micro-electro-mechanical system,opto-mechatronics and so on,the highprecision motion system becomes one of the important components and key technologies.Especially,the piezoelectric-actuated systems designed based on the piezoelectric ceramics play a vital role in the research and applications for micro-nano scale.However,for the position tracking in practical applications,the hysteresis and creep nonlinearity presented in the piezoelectric actuators,and vibrations,friction caused by the mechanical structures affect the final precision significantly.Furthermore,the dexterous and safe interaction between the piezoelectric-actuated systems and environment also poses a great challenge on the design of precision force tracking controller.Therefore,based on the flexure-hinge-guided piezoelectric-actuated system and stick-slip piezoelectricactuated system,from nanometer to micron precision and from feedforward to feedback control,various precision position/ contact force tracking control strategies are studied and designed in this paper,and are applied in the scanning tracking for scanning tunneling microscopes,and myringotomy with tube insertion for treatment of Otitis Media with Effusion.For the motion tracking of triangular waves in raster scanning,although the feedforward iterative learning control is an effective control method,the fixed bandwidth of Q filter in traditional iterative learning control limits the high-bandwidth tracking performance.Therefore,a linear time-varying Q-filter based iterative learning control is proposed in this paper.With the discrete wavelet transform of position errors,the time-varying bandwidth profile is generated based on the modulus maximum of wavelet detail coefficients.Then,through establishing the projection between references and finite-time-response basis function,the error convergence for varying reference during iteration is achieved so that the applications of iterative learning control are extended.The wavelet transform-based method avoids the cross-terms and segmenting manually for general time-frequency decomposition,and the experimental results on flexure-hingeguided piezoelectric-actuated system show that for 20 Hz triangular waves varying during iterations,the proposed controller can achieve flexible and high-bandwidth precision tracking simultaneously.The high-precision and fast tacking for non-raster scanning patterns is an important aspect to improve the imaging speed and quality for scanning tunneling microscopes.For the flexure-hinge-guided piezoelectric-actuated system,a novel signaltransformation based repetitive controller is proposed.Through developing a mapping between the signals and constant amplitude sinusoid/cosine and combining with repetitive control,the controller can provide multiple lower gains at the fundamental frequency and its harmonics so as to ensure zero steady-state errors.The proposed method overcomes the requirement of strictly periodic references or disturbances of conventional repetitive controller.The stability of the control method and convergence of steady-state errors are analyzed with a lifted-system representation respectively for its time-variant property.The experimental results demonstrate that the method has the advantages of a simpler structure and fewer parameters,and the hysteresis,crosscoupling caused errors are removed effectively.For high-speed spiral tracking,the rootmean-square errors and maximum errors are all below 0.065% and 0.18% of the maximum amplitude with the best performance so that the high-speed and high-precision spiral tracking is achieved.Besides,accurate tracking of the interaction force between the piezoelectric-actuated system and the affected human tissues is an important aspect to improve safety and surgery success rate.For the contact phase of myringotomy with tube insertion,an adaptive integral terminal sliding mode force control scheme is presented to achieve precision force tracking when in contact with soft tympanic membrane.An integral terminal sliding manifold based on force error is employed to guarantee the force trajectory tracking with a finite-time convergence.Furthermore,an adaptive control law is also deployed to estimate the controller's parameters and update the switching gain to accommodate the system parameters' uncertainties,and improve the robustness to friction,hysteresis and disturbances,as well as alleviate chattering phenomenon.The stability of the proposed control scheme is analyzed theoretically utilizing methods based upon the Lyapunov theory.Comparative experiments are conducted on the stick-slip piezoelectric-actuated system to verify the effectiveness of the control scheme on S curve and sinusoidal force trajectories tracking with different frequencies when operating in contact with the soft mock membrane.The myringotomy and grommet insertion during the surgical operation requires the high-precision position tracking.A robust adaptive integral backstepping control is formulated to retain high tracking precision.The integral terminal sliding-mode surface is constructed to obtain the property of finite-time convergence,and the second-order auxiliary differential equations are developed to achieve the high-order sliding-mode control-like performance.Through employing integral backstepping methodology and adaptive control,the robustness to external disturbance is guaranteed.The robust stability is proven by the Lyapunov theory.The experiments conducted on the stick-slip piezoelectric-actuated system show that for sinusoidal waves up to 10 Hz,the rootmean-square errors and maximum errors are within 0.65% and 1.35% of the amplitude.For triangular waves,the method also achieves the high-bandwidth tracking.The successful implementation for the surgical operation on mock membrane also indicates the practical value on the application for surgical device.The implementation of the proposed controller is simple without any high-order derivative or observer,and it retains high robustness to hysteresis,friction nonlinearity,unknown disturbance and model uncertainties.Finally,in order to control the position and force simultaneously and retain robust to hysteresis,friction,model uncertainties and so on,a novel robust adaptive impedance control is formulated.The steady-state performance of the target impedance is derived and analyzed through utilizing the nonlinear Hunt-Crossley model.Based on the auxiliary variable containing the impedance error,the integral terminal sliding manifold is proposed to achieve finite-time convergence and improve steady-state tracking performance.Furthermore,an adaptive law is also designed to estimate the upper bound of disturbance to further improve robustness.The stability of the proposed control law is proven by the Lyapunov theory.The experimental results on the piezoelectricactuated system verify that the steady-state analysis based on the Hunt-Crossley model is correct,and the proposed method can also realize precision dynamic force tracking with comparative tracking experiments.Moreover,surgical operation of myringotomy with tube insertion on the mock membrane demonstrates that the achieved force and position precision with the proposed controller can complete the overall processes.In this paper,the position/contact force tracking control strategies are studied systematically at the presence of hysteresis,friction and other disturbance.According to the requirements of different applications,scanning motion tracking and myringotomy with tube insertion,various and suitable feedforward and feedback controllers are designed respectively.The researches in this paper are of great value and significance to the design and practical application of precision motion system in micro-nano scale.