Elsevier

Mechatronics

Volume 75, May 2021, 102511
Mechatronics

Enhanced robust nanopositioning control for an X-Y piezoelectric stage with sensor delays: An infinite dimensional H optimization approach

https://doi.org/10.1016/j.mechatronics.2021.102511Get rights and content

Abstract

In nanopositioning systems with capacitive displacement sensors, the analog-to-digital (A/D) conversion time is not negligible, particularly for the sub-nano resolution cases with 16 (or more) digital bits. The sensor induced delay will complicate modeling and control of such systems, due to the infinite dimensionality of the delay block. In this paper, we analyze the piezoelectric nanopositioner with a Hammerstein-like model incorporating uncertainties and a measurement delay. An infinite dimensional H control approach is proposed for robust and nanopositioning control of the piezoelectric stage. To eliminate the fragility problem of the H controller on its implementation, an equivalent structure including a Finite Impulse Response (FIR) filter is derived, and the matrix-function-based rational approximation (MFRA) for the FIR filter is presented. Real time experiments with the proposed control method are conducted where the positioning performance, robustness, and hysteresis compensation capability are comprehensively evaluated. Comparative studies demonstrate significant improvements over conventional methods such as Proportional–Integral–Derivative (PID) control and the finite dimensional robust control.

Introduction

Piezoelectric stages are widely applied in nano servo applications, such as atomic force microscopy [1] and microinjection system [2]. The nano scale motion of piezoelectric stages can be attributed to the compliant structure driven by piezoelectric actuators [3], [4], [5], while the output displacements are captured by nano resolution sensors. To achieve high-precision servo performances, control methods of such systems have attracted significant research efforts in past decades, where complex model dynamics such as uncertainty and hysteresis were extensively studied [6], [7], [8], [9].

In particular, robust control theory has been well developed to achieve desired performance specifications with robustness against various uncertainties [10], [11], [12], [13], [14], [15], [16]. Note that more and more piezoelectric stages are employing capacitive sensors to measure output displacement, thanks to their sub-nano resolution and capability to measure absolute position zero. However, to achieve sub-nano resolution digital displacement signals for digital servo purpose, ultra-high resolution analog-to-digital (A/D) conversion process, as well as signal filters, is required, which inevitably results in a significant delay in the feedback loop [1], [17]. For systems with delays, the above robust control methods cannot be applied directly due to the infinite dimensionality of the delay.

The presence of delays deteriorates stability margin and makes control design much more complicated [18]. To deal with the infinite dimensionality with engineering applicable structures, various studies on robust control for time-delay systems have been reported, for example, the operator-theoretic method [19], [20] and the J-Spectral factorization method [21]. As an alternative, the Padé approximation based method [17] and the Smith predictor based method [22] were well developed for nanopositioning systems with delays to facilitate the application of finite dimensional control methods [23]. However, these two control schemes do not handle the infinite dimensional delay block directly. The robustness of the both methods has attracted long-lasting discussions [18].

In this paper, we investigate the robust nanopositioning control problem for piezoelectric stages subject to a sensor delay and uncertainties. In contrast to most existing results on finite dimensional strategies, an infinite dimensional H control approach handling the delay block directly is considered. The main contributions of this work are highlighted as follows:

  • 1.

    The Hammerstein-like model of the piezoelectric nano stage is transformed into a linear time-delay system with one lumped uncertainty to facilitate formulating control issues as an infinite dimensional H optimization problem.

  • 2.

    An infinite dimensional H control method is presented with a Finite Impulse Response (FIR) filter based controller structure, where the extension of the infinite dimensional H control theory to non-fragile digital implementation is realized.

  • 3.

    Comparative experiments on the piezoelectric stage are conducted, demonstrating significant improvements on positioning accuracy, transient performance, as well as hysteresis compensation capability.

This paper is organized as follows. In Section 2, the dynamic model of the investigated piezoelectric stage is derived and the positioning control problem for the nanopositioning system with capacitive sensors feedback is briefly discussed. The infinite dimensional H control design is proposed in Section 3. Real time experiments and comparative studies are conducted in Section 4, followed by the conclusion in Section 5.

Section snippets

Modeling of the piezoelectric nanopositioner and problem formulation

We first note that the X-Y nanopositioning stage investigated in this work has a wide range of applications in the precision engineering, such as the atomic force microscope based nano-imaging system [1], [24], the fast tool servo based micro-/nanomachining system [25] and the micro focusing mechanism system [26]. In this section, the time delay property of a piezoelectric stage with capacitive sensors is discussed. A Hammerstein-like model structure including a pure time delay block, as

Infinite dimensional H control design

To optimize the nanopositioning performance with robustnessagainst uncertainties, we investigate the two block H optimization problem for the dynamical model (3): γo:=infCoptstab.P0W1(s)S(s)W2(s)T(s),where S=(1+P0Copt)1 and T=P0Copt(1+P0Copt)1 are the sensitivity and complementary sensitivity functions respectively, and W1(s), W2(s) represent the performance weighting function and the uncertainty weighting function. Note that all the system uncertainties discussed in Section 2.2,

Experimental setup

To validate the effectiveness of the proposed H controller, various real time experiments are conducted on the X-Y piezoelectric stage. As depicted in Fig. 7, the nano-stage is mounted on an air-flotation platform to reduce unnecessary external disturbances. In each axis, a high bandwidth PZT voltage amplifier (with the amplification ratio of 10) is employed to drive the PZT actuator. The overall control system is implemented using MATLAB/Simulink real time control package xPC Target with a

Conclusion

In this paper, an infinite dimensional H control method was developed for nanopositioners considering capacitive sensor induced delays, where an FIR filter based internal loop structure was derived which can be digitally implemented in a non-fragile form by the MFRA for the FIR system. Real time experiments were conducted to comprehensively evaluate the performance of the proposed control approach, where comparative studies on the standard H approach, the PID approach, the μ-synthesis

CRediT authorship contribution statement

Zhiming Zhang: Methodology, Investigation, Data curation, Writing - original draft. Peng Yan: Conceptualization, Supervision, Writing - review & editing, Funding acquisition.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported in part by the Major Basic Research Program of the Natural Science Foundation of Shandong Province, China under Grant ZR2019ZD08, in part by the National Natural Science Foundation of China under Grant 51775319, and in part by the Science and Technology Project of Shenzhen City, China under Grant JCYJ20180305164242690.

Zhiming Zhang received the B.S. degree in mechanical manufacturing and automation from Shandong University, Jinan, China, in 2016. He is currently working toward the Ph.D. degree in mechanical engineering at Shandong University, Jinan, China. His research interests include compliant mechanisms, modeling and control for high precision mechatronic systems.

References (33)

  • KimK.-S. et al.

    A modeling-free inversion-based iterative feedforward control for precision output tracking of linear time-invariant systems

    IEEE/ASME Trans Mechatronics

    (2013)
  • ChenJ. et al.

    UDE-based trajectory tracking control of piezoelectric stages

    IEEE Trans Ind Electron

    (2016)
  • AphaleS.S. et al.

    High-bandwidth control of a piezoelectric nanopositioning stage in the presence of plant uncertainties

    Nanotechnology

    (2008)
  • LiY. et al.

    Design and robust repetitive control of a new parallel-kinematic XY piezostage for micro/nanomanipulation

    IEEE/ASME Trans Mechatronics

    (2012)
  • LiuP. et al.

    Robust antiwindup compensation for high-precision tracking of a piezoelectric nanostage

    IEEE Trans Ind Electron

    (2016)
  • SebastianA. et al.

    Design methodologies for robust nano-positioning

    IEEE Trans Control Syst Technol

    (2005)
  • Cited by (9)

    • A constraint-flow based method of synthesizing XYθ compliant parallel mechanisms with decoupled motion and actuation characteristics

      2022, Mechanism and Machine Theory
      Citation Excerpt :

      Multi-axis compliant mechanisms are excellent candidates to the design of nano-positioning systems and force sensing systems in many modern high-tech field applications from semiconductor manufacture to super-resolution microscopy and bio-nanotechnology [1–9].

    • Microscopic vibration suppression for a high-speed macro-micro manipulator with parameter perturbation

      2022, Mechanical Systems and Signal Processing
      Citation Excerpt :

      The robust stability of the whole system is that the controller can maintain internal stability for a family of perturbation objects based on the nominal system. It has been proved in the literature [36,55,56] that the robust stability of the system with uncertainties depends directly on the vital weight function. Typically, W2(s) denotes the robust weight function for additive perturbation.

    View all citing articles on Scopus

    Zhiming Zhang received the B.S. degree in mechanical manufacturing and automation from Shandong University, Jinan, China, in 2016. He is currently working toward the Ph.D. degree in mechanical engineering at Shandong University, Jinan, China. His research interests include compliant mechanisms, modeling and control for high precision mechatronic systems.

    Peng Yan received the B.S. and M.S. degrees in electrical engineering from Southeast University, Nanjing, China, in 1997 and 1999, respectively, and the Ph.D. degree in electrical engineering from The Ohio State University, Columbus, OH, USA, in 2003. He is currently a full Professor with the School of Mechanical Engineering, Shandong University, Jinan, China. He has worked in various industry positions before joining Shandong University, including as a Staff Scientist at the United Technologies Research Center, East Hartford, CT, USA, from 2010 to 2011, and a Senior Staff Engineer at Seagate Technology, Twin Cities, MN, USA, from 2005 to 2010. His current research interests include design and control of nano-manipulating systems, as well as their applications. He has authored more than 140 scientific papers and filed more than 50 invention disclosures.

    View full text