SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS -PART C: APPLICATIONS AND REVIEWS 1 A survey of modeling and control techniques for Micro-and Nano-electromechanical systems Antoine Ferreira * Sumeet S. Aphale Abstract—In the current times, MEMS and NEMS form a major inter-disciplinary area of research involving science, engineering and technology. A lot of work has been reported in the area of modeling and control of these devices, with the aim of better understanding their behavior and improving their performance. This work presents a review of the emerging advances in the modeling and control of these micro-and nanoscale devices and converges on the exciting research in on- chip control, with a mechatronics and controls perspective and concludes by projecting future trends. Index Terms—MEMS, NEMS, lab-on-a-chip, modeling, control I. INTRODUCTION Though micro-electromechanical systems (MEMS) and nano-electromechanical systems (NEMS) research has gained tremendous popularity and momentum in the past two decades, the potential of small micro-, nano-and even molecular machines was recognized by researchers, especially physicists and chemists almost half a century ago [1]. The race for miniaturizing had begun and finally in 1974, the term Nanotechnology was coined [2]. The development of ‘cluster’ science, [3], and the invention of the Scanning Tunneling Microscope, [4], in the early 1980s ushered the era of nanotechnology and the first book on this subject appeared in 1986, [5]. It is generally accepted that an electrostatically excited tuning fork employing field-effect transistor ”readout” was the first operational MEMS device, [6]. Since then, the MEMS technology has progressed rapidly and in recent years specialized devices for applications such as blood cell separation and analysis are constantly expanding the boundaries of MEMS [7]. NEMS devices have also evolved since their first prototype was successfully demonstrated by researchers at IBM, [8]. Research aimed at developing specific sensor ([9], [10], [11]) and actuator ([12], [13], [14], [15], [16], [17], [18], [19]) technologies for improved MEMS and NEMS devices is ongoing. More details as to the current state-of-the-art for sensors and actuators can be found in [20]. Models that can capture the dynamic behavior of these devices can be of great help in understanding and improving their design and ultimately, their performance. Additionally, as with any dynamic system, a suitable control strategy could Prof. Antoine Ferreira [Corresponding Author] is with the Institut PRISME, ENSI Bourges, 88 Boulevard Lahitolle, 18000 Bourges, France Dr. Sumeet S. Aphale is with the Centre for Applied Dynamics Research align the actual performance of these MEMS/NEMS devices closer to the desired objectives. Therefore, the two key avenues of current engineering interest that have the potential to significantly enhance MEMS/NEMS devices are: (i) modeling and (ii) control. Modeling techniques that lead to a better understanding of these miniature device dynamics are currently being sought after. Accurate dynamic models could lead to specialized control strategies that will in turn lead to major improvements in device performances. In the recent years, a lot of research has been reported in the area of modeling and control of MEMS and NEMS. This paper presents an overview of the emerging innovative modeling techniques applicable to these miniature devices. Different models are presented for system design and control associated with physical mechanisms, geometry/scaling issues or computational aspect for real-time control of MEMS with challenging issues in NEMS. It also reviews the recent advances in the control of MEMS and NEMS devices that have been inspired by the recent innovations in sensors, actuators, modeling techniques and control theory. A. Organization The remainder of this review is organized as follows. Section II presents an overview of the various modeling innovations that describe the behavior of MEMS and NEMS devices. Complexity in modeling is reviewed with respect to associated physical mechanisms, geometry/scaling issues and computational aspect to minimize the real time control issues. This section is further divided into two parts viz: (i) Modeling for MEMS/NEMS design (subsection II-A), (ii) Modeling for MEMS/NEMS control (subsection II-B). Section III will review the various control technique implementations and is divided in to (i) Open-loop control (subsection III-A), (ii) Open-loop control with input pre-shaping (subsection III-B), (iii) Closed-loop control (subsection III-C) and (iv) On-chip control (subsection III-D). Finally, section IV will give the concluding remarks of this review. II. MODELS FOR SYSTEM DESIGN AND CONTROL Today, an abundance of commercial circuit and system simulation tools exist for electronic circuits and control system virtual prototyping. Microelectromechanical systems have been analyzed using the classical physical models or continuum theories for the mechanical (elastostatic or elastodynamic), (CADR), School of Engineering, University of Aberdeen, Aberdeen UK the thermal (thermostatic), magnetic (magnetodynamics) and Copyright (c) 2008 IEEE. Personal use of this material is permitted. the electrical (electrostatic) energy domains [22]. Naturally, However, permission to use this material for any other other purposes must be obtained from the IEEE by sending a request to pubs-permissions@ieee.org the design of reliable actuating techniques requires simple but SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS -PART C: APPLICATIONS AND REVIEWS 2 (a) (b) (c) Fig. 1. Examples of mechanical nodal conventions. F and M are positive valued. (a) Beam in tension, Fx,a = -Fx,b = -F. (b) Beam accelerating in x. Fx,a = Fx,b = F. (c) Moment bending beam with positive curvature in y. Mz,a = -Mz,b = -M, [21]. realistic dynamic models of the device, either in input/output or in the state variable form. Accurate models lead towards optimal system design, better performance, better understanding of the device, short development time, and consequently, lower cost of the device. Furthermore, due to the compact layout, manufacturing tolerance, modeling errors, and environmental changes, MEMS are subjected to parasitics and parameter variations. In order to better guarantee their stability and a certain level of performance, one must take into account these factors in the design of MEMS control systems. This section reviews the models for system design (subsection II-A) and control subsection II-B associated with physical mechanisms, geometry/scaling issues or computational aspect for real-time control of MEMS with challenging issues in NEMS. A. Modeling for MEMS/NEMS design 1) Reduced-order Models: In higher level MEMS/NEMS simulation applications, the computational complexity of getting an output for a given input from the model is simply too high. Thus, model reduction involves reducing the computational complexity of the model by reducing the number of parameters in the original model. If the original model is described by linear ordinary differential equations (ODE) then a typical approach is to write down the algebraic relation in the frequency domain. Reduced order models (ROM) are cheap in terms of memory and computational time and are needed to perform fast and efficient system-level composite circuit for MEMS on-chip development. For practical implementation of feedback control design, the models need to be finite-dimensional. In [25], a reduced nonlinear model was linearized at multiple operating points in order to design a PID-controller tuned via LMI-theory. For MEMS, truncated low-order models can be established this way, using a summation over only selected operating points. In the presence of significant nonlinearities, which often is the case for MEMS, the simple truncated models tend to be too imprecise. However, the technique can be enhanced, by combining structure of the model with finite element analysis a novel way to perform unknown parameters identification. New technique by combining the Taylor series expansion with the Arnoldi method to automatically develop reduced-order models for coupled energy domain nonlinear microelectromechanical devices is given in [26]. Model order reductions via Arnoldi algorithm applied directly to ANSYS finite element models has also been reported [27]. In this work, the authors adopt a micro accelerometer as an example to demonstrate the advantages of this approach. An electrostatically actuated fixed-fixed beam structure with squeeze-film damping effect was examined to illustrate the model-order reduction method in [28]. Compared with the linearized model, these works show that the reduced-order nonlinear models can capture the device dynamic behavior over a much larger range of MEMS operation but stability preservation is not guaranteed and has a low accuracy away from the expansion point. Based on differentiation of the discretized Finite Element (FE) equations for parameterization of MEMS macromodels (see, Figure 1) the authors in [21] computed the governing system matrices as well as high order derivatives (HOD) with regard to design parameters by means of Automatic Differentiation (AD). While the above formalisms were developed primarily for numerical simulations, the possibility to create nonlinear parameterized models based on Karhunen-Loeve decomposition is proposed in [23]. This reduced order model is cheap in terms of memory and computational time and compatible with fast and efficient system-level composite circuit for on-chip feedback control. In the presence of significant nonlinearities, which often is the case for MEMS, the simple linear model order reduction reported in this section tend to be rapidly imprecise due to the vast amount of possible expressions of nonlinearity. General approaches are formulated in the following section for updating the parameters of systems governed by multiphysics equations using advanced optimization techniques. 2) Macromodeling: Several computer algorithms based on 3-D Finite Element Analysis (FEA) have been coupled to 3D design tool to simulate MEMS, [29]. In order to alleviate the computational expense associated with the 3-D analyses, considerable efforts have been devoted to the development of reliable distributed reduced-order models (ROM) for MEMS, [21], [23]. As an illustration, model order reductions via the block Arnoldi algorithm with/without Taylor-series expansion directly to ANSYS finite element models have been proposed for MEMS accelerometers, [27], as well as electrostatically actuated fixed-fixed beam structure with squeeze-film damping effect, [30]. Furthermore, the authors in [31] demonstrated that the resulting ROM can capture the static/dynamic behaviors of the electrostatically actuated MEMS plate very SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS -PART C: APPLICATIONS AND REVIEWS 3 Fig. 2. System level model of an electrostatic torsional actuator. Damping is included as a mixed-level model, (Courtesy of [23]). (a) (b) Fig. 3. (a) ANSYS/Multiphysics finite element model of electrostatic MEMS actuator and (b) SEM micrograph of a fabricated 2 × 2 optical switch MEMS chip based on electrostatic actuation. (Courtesy of [24]). well. Taking the analogy to electronic circuit design further, the next generation of MEMS system designers are starting to use composable MEMS models (macromodeling) [32] as the electrostatic torsional actuator shown in Fig. 2, [33]. It shows a mixed level damping approach where the torsional actuator dynamics is simulated by Navier-Stokes equation- based finite element modeling and the squeeze film damping by lumped-parameter modeling. The pioneering work in forming composable MEMS models is SUGAR from UC Berkeley [34], coventor ARCHITECT, [35], and NODAS from Carnegie Mellon, [36]. ARCHITECT and NODAS use Analog hardware description language (AHDL) descriptions, while SUGAR has its models written in MATLAB. In the later case, the performance of the tunneling MEMS sensor can be estimated and improved based on mechanical-level analysis by ANSYS and system-level analysis by MATLAB, [37]. A feedback control system with one zero and two poles has been synthesized, improving the dynamic range and the bandwidth of the closed-loop system (around 15 kHz). Recently, MEMS design engineers developed a practical method that combines structure of the model with Finite Element Analysis (FEA) in novel way to perform system identification and identify the unknown parameters. The result was a lumped dynamical model of a MEMS device that can be used for the design of feedback control systems, [39]. In principle, any lumped-constant model can be described in this way, thus overcoming the most serious limitation of the equivalent-circuit modeling technique mentioned earlier. A likely reason for the popularity of this technique is that it makes it possible to simulate MEMS using ordinary circuit simulators. An another modeling alternative is to use functional entities representing nanodevices in an object-oriented fashion, termed macromodeling. Macromodeling procedure for coupled-domain MEMS devices with electrostatic and electrothermal effects have been widely presented. Numerical simulation of the dynamics using hybrid BEM/FEM (Boundary Element and Finite Element Method) approach was presented in, [40]. Furthermore, numerical models generated by three-dimensional (3D) finite-element or the finite-difference (FEM/FDM) methods of original coupled-domain MEMS models can be reduced into low-order macromodels (e.g. heat transfer as well as electro-thermal effects) with by Arnoldi-based technique or Krylov-subspace methods [41]. Hybrid analytical/numerical macromodels for the substructures with regular geometry were generated by analytical method and the ones with odd SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS -PART C: APPLICATIONS AND REVIEWS 4 Fig. 4. A table showing the type of physical models, their features and the applications that are most suitable for respective modeling techniques (Courtesy of [38]). geometry by numerical method [42]. These techniques were tested on a generic MEMS device, a microtweezer. The nonlinear tunneling mechanism and electrostatic actuation were linearized using small-signal approximation. It must be noted that exporting macromodels for MEMS simulation requires the interfacing of various commercial tools for CAD (e.g., SolidWorksTM ), FEA, simulation of electronic circuits (e.g., AHDL/VHDL language), control systems (e.g., Matlab/SimulinkTM ), multibody systems (e.g., ANSYS/MultiphysicsTM ) and also the microfabrication processes. Figure 3(a) illustrates the microfabrication of an electrostatic MEMS comb-drive actuator [24] capable of moving the MEMS mirror 50 µm designed using ANSYS/Multiphysics finite element model (Fig. 3(a)). This simulation can solve coupled field problems such as electrostatic forces acting on the mechanical structures, and is also capable of performing contact analysis when the glider contact the mechanical stop. There are definite drawbacks, the simulation of HDL models or models written in other high-level languages is usually considerably slower than the simulation of equivalent models built into the simulator. Furthermore, it is noteworthy to discuss macromodeling applicability in conjunction with MEMS control design since realtime feedback control issues are still unsolved. 3) Multiscale Models: The ability to design reliable MEMS/NEMS devices demand new simulation capabilities due to the length and time scaling effects at nanoscale [43]. Combination of classical microforces phenomena with quantum fields and molecular considerations become key issues to the point that thermal fluctuation influences the NEMS operation. Furthermore, the roles of surface and defects become more dominant. Finally, the behavior of materials at nanometer scale begins to be atomistic rather than continuous. Taken together, it gives rise to anomalous and often nonlinear effects, i.e., nanomechanics (Casimir effect, van der Waals, charges quantization), nano-optics (charge transfer), electrostatic-fluidics effects (dielectrophoresis, electro-welting, electroosmosis), nanomagnetics (paramagnetism), and so on. The challenge now faced by NEMS designers is to bridge the different scales to a more general framework, which has been coined as multiscale modeling [44]. Conceptually, two categories of multiscale simulations can be envisioned: both sequential and concurrent. (i) Sequential multiscale simulations The sequential methodology attempts to piece together a hierarchy of computational approaches in which large-scales models use the coarse-grained representations from more detailed smaller-scale models. In doing so, the simulations are running independently of each other and a complete separation of both length and time scales are achieved [80], [76]. Some examples of sequential coupling show that to accurately model MEMS/NEMS devices at least three length-scales need to be explored: mesoscopic at the package level; microscopic at the actuator/sensor level and nanoscopic at the material level. Reliability of packaged polysilicon microelectromechanical systems involves the computational study of environmental effects to predict the long-term performance of MEMS packages at mesoscopic and microscopic length scales. The authors in [75] present a SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS -PART C: APPLICATIONS AND REVIEWS 5 TABLE I COMPARISON OF MODELING AND SIMULATION APPROACHES FOR MEMS AND NEMS DESIGN. Length scale Modeling Key Ref. Time scale Computational complexity (.) Computational error (%) Modeling level L = 10µm Macroscopic Classical Physics Physical models 1s = t = 10s O(n2 ) ~ 10-30 % System-level [45],[46],[47] [48],[49],[50] Dynamical State-Space model [51],[52],[53] Lumped Dynamical model [54], [55], [56] High-Order Derivatives model [57],[58] Composite circuit macromodels VHDL/AHDL language Mixed-level model [34],[35],[36] [42],[40],[59] O(n2 ) ~ 20 % System-level Continuum Models Finite Element Methods Boundary Element Methods [60],[21] [61],[62] O(n3 ) ~ 15 % MEMS part-level Model Order Reduction Krylov algorithms Arnoldi algorithms [23],[63] [26],[30],[64] O(n2 ) ~ 5-10 % MEMS part-level Mesoscopic 100nm = l = 1µm Stochastic Methods Direct Monte Carlo Methods Kinetic Monte Carlo [65] [66] 1µs = t = 1ms O(n3 ) ~ 25% Functional-level Functional-level 10nm = l = 100nm Molecular Molecular Dynamics [67],[68],[69] 1ns = t = 1µs O(n4 ) ~ 20-30 % Functional-level Tight-Binding Molecular Dynamics [70] Functional-level Coarse-Grained Molecular Dynamics [71] Functional-level Stochastic Dynamics [72] Functional-level Atomistic 1 °A = l = 1nm Density Functional Theory Hartree-Fock Approximations [73] [70] 1ps = t = 1ns O(n5 ) ~ 15 % Atomic-level Atomic-level 1 °A = l = 100µm Multiscale Continuum/MD coupled models [74] 1ps = t = 10s O(n5 ) ~ 5-10 % System-level FEM/CGMD coupled models [71] MEMS part-level FEM/MC coupled models [75][76],[77] System-level continuum/MD/QM coupled models [78],[79] System-level . where n is the number of features in environment. multiscale finite element modeling (FEM) approach coupled to Monte-Carlo (MC) analysis for MEMS failure prediction. In a same way, a predictive-science-based multiscale modeling and simulation platform is proposed in [81] to predict material performance issues, such as radiation, thermo-mechanical cycling and damage and fracture due to shocks. The computational coupling of the atomic-scale description of nanomaterials (Molecular Dynamics (MD) simulation) to microscale actuators designs (traditional FDM/FEM) pose severe challenges. MD simulation cannot simulate the whole systems due to its prohibitive computational cost, whereas continuum FEM/BEM scales poorly with system size and only approximately account for effects at material interfaces. To remedy these inadequacies, several authors coupled FDM/BEM simulations to MD models whose underlying physics are derived from nanomechanics theory [53], [82], [83], nanoelectronic structure theory [62], nanofluidics theory [84], and molecular biology [77]. In overall, the sequential multiscale model showed good qualitative agreement with the experimental measurements but requires more refinement to achieve good quantitative agreement. (ii) Concurrent multiscale simulations The concurrent multiscale approach attempt to link methods appropriate at each scale together in a combined model, where the different scales of the system are considered concurrently and communicate with a hand-shake procedure. The literature contains numerous methods of concurrent coupling; (i) the combined finite element atomistic method (FEAt), (ii) the material point method (MPM), (iii) the local quasicontinuum method (QC), (iv) the bridging scale method, (v) the atomic-scale finite element method (AFEM), and (vi) coarse grained molecular dynamics (CGMD) [85], [43]. Molecular dynamics simulations are commonly used to investigate size-dependence of the elastic properties of the nano-scale silicon cantilevers [69]. It reveals that continuum mechanics modeling can still be used on nanoscale structures provided that the dependence of elastic constants on dimensional scaling is accounted for. At a larger scale Coarse-Grained Molecular Dynamics (CGMD) modeling SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS have been developed [71] to describe the behavior of the mechanical components of MEMS down to the atomic scale. It builds a generalized finite element formalism from the underlying atomistic physics in order to ensure a smooth coupling between regions governed by different length scales. Various electrostatic models namely: the classical conductor model [74], the semiclassical model [86], and the quantum- mechanical model [53], are being used for electrostatic analysis of NEMS at various length scales. The design methodology facilitates, under restricted conditions, the insertion of quantum corrections to nano-scale device models, during simulation. In the case of NEMS-based electrostatic actuation, Figure 4 shows the evolution of modeling theory w.r.t. device length scale : from classical continuum models to atomistic quantum mechanical models. In [38], a multiscale method, seamlessly combining semiclassical, effective-mass Schro¨dinger, and Tight-Binding Theories (TBT), is proposed for electrostatic analysis of silicon nanoelectromechanical systems. In [87], an integrated modeling methodology for nano-scale electronic devices has been proposed. This methodology includes domain-oriented approximations from ab-initio modeling and the selection of quantum mechanical compact models that can be integrated with basic electronic circuit or non-electronic lumped-element models. Finally, molecular dynamics (MD) and ab-initio quantum mechanics(QM) coupled to virtual reality (VR) techniques have been developed in [78], [79] for the prototyping of biological NEMS. The operator can design and characterize through molecular dynamics simulation, the behavior of bio-nanorobotic components and structures through 3-D visualization. In these works, the nonlinear continuum elastic theory, with material properties extracted from MD simulations, is combined with either the classical, semiclassical, or the quantum-mechanical electrostatic model and the continuum theory for the van der Waals energy domain to compute the self-consistent electromechanical behavior of biological NEMS. From the point of view of control, the concurrent coupling between the mechanical and the electrical energy domains at nanoscale necessitates a proper understanding of relevant physical theories for NEMS feedback control [88]. Actually, carbon nanotube-based NEMS devices (nanoswitch, nanotweezers) are actuated using analytical energy-based methods modeling (electrical capacitance model including van der Waals forces as well as finite kinematics) to predict the structural behavior and instability of the on/off states of the nanoswitch, or the open/close states of nanotweezers [61]. Recently, the influence of control parameters on the stationary oscillations of carbon nanotube-based oscillators via molecular dynamics simulations have been conducted [67]. The control of oscillator motion in the case of considerable fluctuations through the control force has been rendered possible. The methodologies reported here are completely general and as such are expected to be useful in the optimal control of nanotube-based NEMS devices. Table 1 gives a comparison -PART C: APPLICATIONS AND REVIEWS table of various modeling technologies with its pros and cons. Methods used for simulation of several properties of MEMS/NEMS differ in their level of accuracy and in the computation time necessary to perform such calculations. Accordingly, the time scales that each of these methods can handle can be from a single total energy for the most accurate calculations, to picoseconds for ab-initio molecular dynamics simulations, and up to seconds for classical physics. B. Modeling for MEMS/NEMS control 1) Physical Models: Physicals methods for determining lumped dynamical models of thermal, piezoelectric, magnetic or electrostatic MEMS and NEMS devices for purposes of feedback control have been studied extensively in literature. Current modeling works are mainly focused on the empirical responses of the system dynamics, black-box models, as a practical model for real-time control, but offer minimal insight into the governing equations. System identification based on measured sets of input and output data obtained from exciting the system with pseudo random binary data (PRBS) gives a good fit to the measured data. The MEMS dynamics are dominated mainly by the first mode which can be accurately modeled by a mass-spring-damper second order-model, e.g. piezoelectric MEMS scanner [47], polymer MEMS actuators [50], piezoelectric microrobot-onchip [89] and electrostatic MEMS vibrational gyroscope [90]. However, when the number of parameters grows, it becomes more difficult to span the complete parameter space, since each parameter lets the number of possible variations grow in an exponential way. As example, the fast dynamics of MEMS systems require higher-order models leading to complicated model-based controllers. As a more detailed approach, the gray-box models are developed for determining lumped dynamical models of MEMS devices, [54], [55], [56], for purposes of feedback control. A model consisting of millions of equations (e.g., a FEM model) is surely more difficult to handle and takes more time to solve than an analytic expression based on a simplified gray-box model. In [91], the authors determined a dynamical state-space model for control of thermal MEMS devices. The importance of temperature-dependent parameters was emphasized for dynamical modeling for purposes of feedback control. In [58], a computationally efficient model was developed for investigating the dynamics of the voltage-driven MEMS device embedded in a dielectric fluid. However, these models were partly based on physical principles while also relying on empirical results to define complex physical processes. Due to the compact layout [92], manufacturing tolerance [93], modeling errors [22], and environmental changes (e.g., adhesive surface interactions, and scale dependent material and thermal properties) [94], MEMS devices are subjected to parasitics and parameter variations. In order to better guarantee their stability and a certain level of performance, one must take into account SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS these factors in the design of MEMS control systems. In the most complex form, white-box models with partial differential equations (PDEs), e.g., [57],[58], attempt to explain the underlying physics for the sensing and actuation responses of MEMS and NEMS. Nonlinear models based on finite-difference discretization of MEMS structures, e.g. lateral electrostatically-actuated DC-contact MEMS [55], and applying boundary conditions have been recently solved using a Gauss-Seidel relaxation iteration scheme. More efficiently and equally accurate during circuit simulation than PDEs, Volterra-series-based modeling describes the frequency dependence (e.g., the mechanical resonance) in combination with the nonlinear behavior of the MEMS variable capacitor [95]. As the complexity of such models involves model reduction techniques, there is always a tradeoff between accuracy of the model or possible range of application. 2) Advanced Modeling Algorithms: Recently, black-box advanced modeling algorithms of non-electronic parts has been introduced in MEMS modeling, so enabling radically faster simulation without concurrent algorithms and parallel computation, e.g. artificial neural networks (ANN), genetic algorithm (GA) optimization, model prediction (MP), or fuzzy logic algorithms (FL). In [96], a lumped model of the capacitive transducer, being the part of a MEMS capacitive pressure sensing system, was created using an ANN. The ANNs here are considered universal approximators, convenient for black- box device modeling. A general approach was formulated in [31] for updating the parameters of systems governed by multiphysics equations using an optimization technique based on Genetic Algorithms (GAs). This approach was demonstrated on a MEMS micromirror which was governed by both structural and electrostatic physics. For systems with fast dynamics such as those in MEMS, a hardware embedded real- time implementation of model predictive control (MPC) has been investigated in [97]. The results show that MPC would be an appropriate controller implementation since the size and the application precludes the use of a dedicated computer. Finally, a method for reliability prediction was presented in [98], based on a combined fuzzy-logic and physics-of-failure approach. The specific case of a MEMS Fabry-Perot interferometer was analyzed and the failure rate estimations are discussed. Similar fuzzy logic control algorithms have been applied to optimally charge the microbattery of on-board MEMS sensors [99]. Recent manufacturing advances have opened the path for the fabrication of micromechanical devices and electronic subsystems under the same manufacturing and packaging process, thereby opening the path for the use of advanced modeling algorithms towards systems-on-chip applications. III. CONTROL SCHEMES Presence of sensor dynamics, fast high-frequency system dynamics and extremely sensitive system parameters make the control of MEMS devices a complex task. Over the years, researchers all over the world have come up with feasible control algorithms for MEMS devices. Based on -PART C: APPLICATIONS AND REVIEWS these results, control techniques for MEMS can be grouped under three broad classes viz: open-loop control, open-loop control with input pre-shaping and closed-loop control [100], see Fig. 5. The choice of the control technique depends on various factors such as application, needed electronic circuitry, device dynamics, space constraints and sensor availability / implementation. The following sub-sections will review the various control strategies mentioned earlier. The section will close with a review of the on-chip control strategies that have been implemented by researchers so far, (Subsection III-D). A. Open-loop control During the infancy stages of MEMS technology, most MEMS devices were controlled in open-loop by applying very simple control inputs. This was mainly due to the relatively high speed of actuation as well as the inability of the then existing sensor technologies to procure noise-free sensory information that was unbiased by the sensor dynamics. Though advances in sensor and actuator technologies have further pushed the boundaries of accurate sensing at the micro-and nano-levels, successful integration of these sensors in MEMS remains an ongoing challenge, [20]. Recent advances have resulted in improving the traditional MEMS designs to achieve better dynamic performance under open- loop actuations, [28]. Open-loop control for large deflection electrostatic actuators was reported in [17]. In this paper the authors incorporated significant design improvements to the existing comb-drives designs [101], [102]. These improvements included reducing the actuator area by half, redesigning comb-teeth and suspensions to reduce side instability and using a launch and capture actuation scheme. MEMS deformable mirrors have been popularly controlled in open-loop, [103]. The open-loop scheme delivered accurate tracking to within 3% error. Wavelength-division multiplexed (WDM) routers with analog micromirror arrays were shown to operate in open- loop with excellent repeatability and stability, [104]. High repeatability and long-term stability of a MEMS wavelength selector switch in open-loop operation was demonstrated in [105], though it lacked a 100% add / drop functionality. A low-drift micromirror in open-loop control was demonstrated in [106]. Open-loop control of a MEMS deformable mirror using a nonlinearly constrained quadratic optimization approach has also shown improvements in performance, [107]. In this case, with an a-priori knowledge about the aberrations in the target waveform, a quasi-steady state control was obtained. Though simulated results reportedly showed improved performance, practical implementation of this rigorously mathematical technique may be quite challenging. Recently, MEMS actuator designs are being modified to give better open-loop performance, [108]. An improved modeling technique that resulted in open-loop control of a tunneling accelerometer for very high resolution acceleration measurement was reported in [109]. In this case the tunnelling accelerometer was modeled based on a clamped micro-circular plate with a tunneling tip and the SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS -PART C: APPLICATIONS AND REVIEWS 8 (a) (b) (c) Fig. 5. (a) Scanning electron micrograph of a MEMS variable optical attenuator (VOA). (b) Schematic of the experimental setup of the VOA(c) Open-loop step responses for the simulated model and the actual VOA, (Courtesy of [100]). classic Kirchhoff thin plate theory was used for deriving the governing equations. A new hardware platform for tuning a MEMS based gyroscope in open-loop by measuring the frequency response of the device was reported in [110]. These platforms tuned the gyroscopes based on an evolutionary computational technique that improved the sensitivity of the gyroscopes and also enabled closed-loop operation. An open-loop technique to address the Sagnac effect in a fiber-optic gyroscope based on MEMS/NEMS fabrication has also been proposed in [25]. As problems such as inherent system nonlinearity, induced vibrations and effects such as stiction and friction cannot be completely addressed using open-loop control in many MEMS devices, input pre-shaping was seen as the next logical s tep in MEMS control. B. Open-loop control with input pre-shaping This technique relies on the fact that the static and dynamic behavior of many MEMS devices can be accurately modeled and in most cases, linearized. In this technique, the input signals are made more complex by shaping them in a way such that the adverse effects of the system dynamics are minimized (Ex: Bandlimiting the trajectory signal such that the natural frequencies/system resonances were not excited), [112]. For input pre-shaping, an accurate dynamic model of the system is of paramount importance, if any performance improvement is expected. An open-loop method that predicted control voltages generating prescribed surface shapes on a MEMS deformable mirror was given in [113]. In this work, an analytic elastic model was used for the mirror membrane and an empirical electromechanical model was used for the actuator dynamics. Open-loop control with input pre-shaping has also been applied to control oscillations of MEMS based gyroscopes. For accurate angular rate measurement, the drive mode oscillation amplitude of the second mass has to be kept constant. By approximating the gyroscope by a lumped mass-spring-damper model and applying pre-computed actuation voltages, the oscillation amplitude can be kept constant as shown in , [114]. For MEMS devices that involve multiple moving parts, such as MEMS mirror arrays, a feed-forward based control has been patented, [115]. This patent was specific to MEMS based, optical mirror arrays where motion of an active mirror has an aerodynamically disturbing effect on the neighboring static mirrors in the array. In this technique, feed-forward control signals with a normalized profile that minimized the aerodynamic coupling between the static mirrors were employed to cancel the induced disturbances. Feedforward control of a MEMS optical switch was reported in [116]. In this implementation, feed-forward was used to force the switch to reach the desired position in a fast and accurate manner with minimal overshoot. Very recently, a patent was awarded for a input shaping actuation technique for MEMS devices, [117]. In this patent, a filtered voltage signal shaping technique has been demonstrated. This scheme is mainly useful in conjunction with MEMS devices that have micro-cantilevers and other vibrating elements whose natural resonances are minimally damped. The patent is based on results obtained in [118] by actuating a two-axis gimbal-less scanner using the open-loop with input pre-shaping technique presented in the patent. Filtering the input voltages may not always be feasible as it adds to either the system or the computation cost. Additionally, sub-optimal filtering may lead to unachievable slew-rates and supply saturation. It is the property of electrostatic MEMS actuators to generate a residual charge in their insulating layer that results in sticking of the electrode and increases response time. To prevent this sticking of electrostatic MEMS actuators and generate fast actuator response, an input pre-shaping technique was described in [119]. The patented technique of Input Shaping was demonstrated to potentially nullify unwanted vibrations in MEMS devices such as MEMS optical switches in [120]. A similar technique was used in [111], to drive a micromirror to a desired tilt angle without residual vibrations, see Figure 6. The key advance in this input shaping technique was the inclusion of nonlinear system behavior, thus making it suitable for application in conjunction with a wide range of MEMS systems. The inherent reliance of the input-preshaping technique on an accurate system model as well as a-priori information of the system behaviour limits the adaptability and robustness that SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS -PART C: APPLICATIONS AND REVIEWS 9 (a) (b) Fig. 6. (a) Controlled (using shaped signal -solid) and uncontrolled (using standard step commands -dashed) tilt angle responses achieved by the MEMS micromirror. (b) Corresponding shaped (solid) and unshaped (dashed) actuation voltage signals, (Courtesy of [111]). can be built into this particular control technique. Finally, the combined advances in MEMS technology, sensor and actuator designs, system analysis tools and the ever-present demand to push the boundaries of performance in terms of speed, reliability and accuracy have led to the MEMS system be controlled by employing complicated closed-loop strategies [121]. C. Closed-loop control Standard control techniques such as PID have been implemented on MEMS devices manufactured in bulk, such as MEMS based sensors and switches [123], [124], see Fig. 7. MEMS based sensors have been using closed- loop control for quite some time. Hitachi demonstrated a MEMS based closed-loop silicon accelerometer more than a decade ago, [125]. Closed-loop control was used for a MEMS micro-cantilever based pressure sensor [126]. In this application, an electromagnetic beam integrated onto a standard silicon pressure sensor diaphragm was driven to resonance using closed-loop control. As the diaphragm deflects under pressure, the stress in the beam caused a change in its resonant frequency. This change was found to be a highly sensitive measure of pressure [127]. This device offered wide dynamic range, high sensitivity, and high stability. It was also easy to be interfaced with digital compensation circuitry. Another successful application of closed-loop control in MEMS was reported in [128], where improving measurement accuracy was the main objective. Feedback control has been employed to accurately regulate the gap distance in an electrostatic MEMS based Fabry-Perot interferometer, [129]. In this implementation, a feedback circuit capable of sensing the property of the active device and providing an electrical stop when the minimum separation distance was achieved was integrated. Closed-loop feedback control has been a common strategy to correct for machining imperfections in MEMS based gyroscopes. [130], [114] proposed active nonlinear and adaptive drive control approaches to compensate for errors due to device imperfections. Closed-loop tuning of a MEMS based gyroscope was reported in [110]. A custom-built integrated circuit that manages the signal filtering and provides real-time control for the JPL-Boeing manufactured MEMS based gyroscopes was reported in [131]. This technique used an ASIC that enabled the gyroscope to reject vibration disturbances and damped the transfer function by almost 40 dB. A US Patent for an application specific integrated circuit capable of exciting a selected gyroscope mode, induce damping and demodulate the signal containing the angular rate information to in-phase and quadrature components was issued, [132]. This circuit featured attractive properties such as low power consumption as well as ease of sensor integration. A dual-stage control algorithm that provided on-site identification of imperfections based on the dynamic response of the device and compensated for it using nonlinear electrostatic parallel plate actuators was proposed in [133]. In this paper, the authors first showed that using feedback alone to compensate for large structural imperfections (to the tune of 10%) would seriously compromise the device performance. Consequently they successfully employed a feedforward control loop to reduce large imperfections and combined it with a feedback loop to compensate for the device non-idealities and perturbations. [134] presented a novel architecture for the digital control of MEMS based gyros. Digital control was also proposed for performance optimization of a MEMS based gyroscope, [135]. In general digital control was shown to offer more flexibility in terms of algorithms as well as control parameters. FPGAs used in these implementations significantly speed up the development process due to their ease of programmability. Cross-coupling SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS -PART C: APPLICATIONS AND REVIEWS 10 (a) (b) Fig. 7. (a) Implementation of the Adaptive Gain Control (AGC) feedback system for the velocity-controlled gyroscope. (b) Magnitude of the driving signal and envelope of the associated velocity signal under various vacuum conditions for open-loop (a) and closed-loop (b), (Courtesy of [122]). and fabrication imperfections are the major performance limiting factors in MEMS based gyroscopes. Adaptive control based on velocity estimation has shown promise in alleviating these problems and improve the overall performance of the gyroscope by achieving larger operational bandwidth, eliminating zero-rate output, enabling self-calibration and deeming the gyroscope highly robust to parameter variations, [136], [137]. To further improve the gyroscope performance by accurately estimating the unknown angular velocity, sliding mode control has also been formulated, [138], [139]. These investigations proved that though computationally intensive, both these nonlinear approaches could significantly enhance the device performance. Furthermore, they also showed that sliding mode control of the vibrating proof mass resulted in a better estimate of the unknown angular velocity than that of the model reference adaptive feedback controller. An active disturbance rejection control scheme was proposed recently to address issues such as mechanical-thermal noise, parameter variations, quadrature errors and the mismatch of natural frequencies between two axes, [140]. In this work the two main control problems addressed were the vibrating modes of the gyroscope axes and the time-varying rotation rate estimation. These major issues can also be alleviated using an adaptive control method based on Lyapunov functions, as demonstrated in [141]. A discrete time observer-based adaptive control algorithm for improved angular rate estimates has also been reported, [142]. Performance enhancement for a probe-based data storage system was reported in [143]. In this paper, the authors proposed a position control system that resulted in accurate positioning of the micro-cantilever probe over a particular SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS sector of the data storage disc. Positive Position Feedback (PPF) control was implemented successfully to provide active damping to a piezoelectric MEMS acoustic sensor, [144]. A detailed comparison between open-and closed- loop control of a MEMS electrostatic comb drive was given in [39]. Model Reference Adaptive Control (MRAC) technique was formulated for tracking control of MEMS based comb resonators, [145]. A real-time implementation of this technique demonstrated its ability to handle multiple uncertainties in device parameters that occur due to machining imperfections. In [116], a linear feedback controller was used to shape the system dynamics in a MEMS optical switch, resulting in fast switching operation. Nonlinear control was also used in manipulating MEMS based mirrors for high tilt and pointing accuracy, [146]. In this application, digital implementation of a full-state feedback was carried out resulting in a substantial increase in the mirror’s angular operation range and a reduction in the long-term angular noise. Nonlinear sliding mode control applied to controlling the position of a lateral comb resonator has been simulated in [147]. A cooperative angle control scheme to reduce the output stable control time in MEMS optical switches was proposed in [148]. In one of the novel applications, feedback control has been employed to provide accurate input gains and implement signal up-modulation to a MEMS based high-performance operational amplifier, [149]. In this application, the input stage of the operational amplifier is a MEMS based variable capacitor that converts low-frequency input voltages to high-frequency AC currents, resulting in reduced offsets and low-frequency noise. In most closed-loop control of MEMS devices, the control loop was implemented using external circuitry and computing facilities. In many cases, even the sensors were independent and not an integral part of the MEMS device. With improved fabrication methods, component densities on a chip have increased drastically and on-board sensors and power sources have become the norm, [150], [151], [152], [153]. Thus, the system-on-chip concept with on-chip control is now gaining popularity [154], [155], [156], see Fig. 8. D. On-chip feedback control: The current trend The main advantages of on-chip feedback are: (1) improved linearity, (2) improved signal-to-noise ratio and (3) improved accuracy due to ease of compensation for interferences and system dynamics. The vast improvements in MEMS design and fabrication have led to real-time on-chip feedback control being the current thrust of many research endeavors. One of the main control issues for on-chip applications is power generation. Electrical power is needed for the actuation as well as sensory systems used in MEMS. Generating this power efficiently with-in the given space constraints of a MEMS chip, without causing insulation, dynamics or interference problems is a major concern and research focus. Some on-chip power sources have been reported and various power generation schemes are being investigated, [152], [151]. -PART C: APPLICATIONS AND REVIEWS Microfluidics is an area where real-time feedback has been applied successfully, [158]. Precise handling of microfluidics in continuous-flow by using a flow sensor to monitor and on-chip pumps with feedback control to regulate the flowrate was presented in [159]. Electrowetting-on-dielectric (EWOD) has been proposed as a method of actuation for on-chip droplet generation [160]. Due to its compatibility with miniaturization, simple device configuration and fabrication, capacity to generate large forces at microscale, and low energy consumption EWOD has gained popularity in microfluidic applications. To monitor droplet volume and control applied voltages for on-chip droplet generation of constant volumes real-time feedback control was necessary. A successful feedback strategy that resulted in automated volume-controlled on-chip generation of droplets was reported in [161]. To account for the uncertainties during droplet separation, an improved feedback scheme was proposed in [162]. In this work the authors combined voltage modulation, capacitance sensing and a discrete-time PID controller to obtain significant improvements in droplet volume uniformity when compared to open-loop and standard closed-loop techniques. On-chip control has also been applied to MEMS based high-speed synchronous micromotors, [22]. In this work, on- chip VLSI drivers are used for various signal processing, filtering, computing, interfacing and amplification tasks and the control of micromotors is achieved by applying the proper phase voltages to the micromotor windings. The control technique also incorporates robust tracking and disturbance rejection. Nonlinear control of electrostatic MEMS using a novel integrated charge and position sensor was reported in [163]. This technique resulted in full gap operation and improved transient performance. The control technique showed on-chip implementation potential as it could use the local integrated circuit components and the required sensor was easy to fabricate, did not increase device footprint and had negligible effect on the device dynamics. The design of an on-chip CMOS potentiostat was reported recently, [164]. The potentiostat was mainly developed for controlling the volume of conjugated polymer film used in microactuators. This on- chip mechanism was proposed for controlling microactuators used in cell capture microsystems. With the advent of nanoelectromechanical systems (NEMS), on-chip sensor technologies are being revolutionized, [157], [165]. As a result, the future is bright and holds exciting prospects for on-chip control of MEMS devices. IV. CONCLUDING REMARKS This review tries to be complete in its general scope by reviewing most major works in the field of MEMS/NEMS modeling and control. Though it is impossible (due to the size and time constraints) to list every single work in a review of such a varied and dynamic field, the authors believe that this paper provides the reader with the most up-to-date information about the various advances that have taken place over the years in the modeling and control of MEMS/NEMS devices. SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS -PART C: APPLICATIONS AND REVIEWS 12 (a) (b) (c) Fig. 8. (a) Scanning electron micrograph of the released microgrippers showing the on-chip photo detectors placed beneath the gripping sites. (b) Schematic of the photo-detector circuit (c) Gripping of a polystyrene bead: before (top); after (bottom), (Courtesy of [157]). With the advent of better, faster computing hardware and dedicated software, it will be prudent to say that the field of MEMS/NEMS is bound to see an even greater influx of academic and industrial interest. As the fields of physics, chemistry, biology and mathematics evolve and fuse together, more realistic models that capture the behavior of these micro- scale systems most accurately should be a key result. This, in turn, will combine with the fast developments occurring in the areas of very high device density chip fabrications and flexible electronics to produce control techniques that will make the desirable performance of a MEMS/NEMS device easily realizable, robust and adaptive. ACKNOWLEDGMENTS This work was supported by the French National Agency (ANR) PIANHO project ”Innovative Haptic Instrumental Platform for 3D Nanomanipulation” under the project number ANR-09-NANO-042-02. REFERENCES [1] R. Feynman, “There’s plenty of room at the bottom,” Lecture at the American Physical Society meeting at Caltech on 29th December, 1959. [2] N. Taniguchi, “On the basic concept of ‘nano-technology’,” Proceedings of the International Conference of Production Engineering (Japan Society of Precision Engineering and International Institution for Production Engineering Research), Tokyo Part II, 1974. [3] H. W. Kroto, J. R. Heath, S. C. O’Brien, R. F. Curl, and R. E. Smalley, “C60 : Buckminsterfullerene,” Nature, vol. 318, pp. 162 – 163, 1985. [4] G. Binnig and H. Rohrer, “Scanning tunneling microscopy,” IBM Journal of Research and Development, vol. 30, no. 4, 1986. [5] K. E. Drexler, Engines of Creation: The Coming Era of Nanotechnology. Doubleday, 1986. [6] H. C. Nathanson, W. E. Newell, R. A. Wickstrom, and J. R. Davis, “The resonant gate transistor,” IEEE Transactions on Electronic Devices, vol. 14, pp. 117 – 133, 1967. [7] X. Chen, D. F. Cui, C. C. Liua, and H. Lia, “Microfluidic chip for blood cell separation and collection based on crossflow filtration,” Sensors and Actuators B: Chemical, vol. 130, pp. 216 – 221, 2008. [8] M. Despont, J. Brugger, U. Drechsler, U. Drig, W. Hberle, M. Lutwyche, H. Rothuizen, R. Stutz, R. Widmer, G. Binnig, H. Rohrer, and P. Vettiger, “VLSI-NEMS chip for parallel afm data storage,” Sensors and Actuators A: Physical, vol. 80 (2), pp. 100 – 107, 2000. [9] H. Yamaguchi, S. Miyashita, and Y. Hirayama, “InAs/AlGaSb heterostructure displacement sensors for MEMS/NEMS applications,” Proceedings of the International Conference on Molecular Beam Epitaxy, vol. 251, pp. 556 – 559, 2003. [10] V. Gayathri and R. Geetha, “Carbon nanotube as NEMS sensor effect of chirality and stone-wales defect intend,” Journal of Physics: Conference Series, vol. 34, pp. 824 – 828, 2006. [11] X. Li, J. Han, H. Bao, and Z. Yang, “Integrated spm probes with NEMS technology,” Sensors and Actuators A: Physical, vol. 133 (2), pp. 383 – 387, 2007. [12] J.-H. Park, C.-H. Lee, Y.-H. Park, Y.-D. Kim, C.-H. Ji, J. Bu, and H.-J. Nam, “A fully wafer-level packaged RF MEMS switch with low actuation voltage using a piezoelectric actuator,” Journal of Micromechanics and Microengineering, vol. 16, pp. 2281 – 2286, 2006. [13] I. Kanno, S. Tsuda, and H. Kotera, “High-density piezoelectric actuator array for MEMS deformable mirrors composed of PZT thin films,” IEEE/LEOS International Conference on Optical MEMs and Nanophotonics, (Freiburg, Germany) 11 -14 August, pp. 132 – 133, 2008. [14] P. Srinivasan and M. S. Spearing, “Optimal materials selection for bimaterial piezoelectric microactuators,” IEEE Journal of Microelectromechanical Systems, vol. 17, no. 2, pp. 462 – 472, April 2008. [15] S. Heo and Y. Y. Kim, “Optimal design and fabrication of MEMS rotary thermal actuators,” Journal of Micromechanics and Microengineering, vol. 17, pp. 2241 – 2247, 2007. [16] G.-K. Lau, J. F. L. Goosen, F. van Keulen, T. C. Duc, and P. M. Sarro, “Polymeric thermal microactuator with embedded silicon skeleton: Part idesign and analysis,” IEEE Journal of Microelectromechanical Systems, vol. 17, no. 4, pp. 809 – 822, August 2008. [17] J. D. Grade, H. Jerman, and T. W. Kenny, “Design of large deflection electrostatic actuators,” IEEE Journal of Microelectromechanical Systems, vol. 12, pp. 335 – 343, 2003. [18] R. Kornbluh, “Dielectric elastomer actuators: muscles that morph,” ICASE Lecture Series on Morphing (Hampton, VA: NASA Langley Research Center), 2001. [19] T. Tanaka, K. Saeki, and K. Matsuki, “Dynamic characteristics of high field electro-active silicone and acrylic elastomer actuator devices,” IEEE Conference on Electrical Insulation and Dielectric Phenomena, (Kansas City, Missouri, USA), 15 -18 October,, pp. 202 – 205, 2006. [20] D. J. Bell, T. J. Lu, N. A. Fleck, and S. M. Spearing, “Mems actuators and sensors: observations on their performance and selection SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS -PART C: APPLICATIONS AND REVIEWS 13 for purpose,” Journal of Micromechanics and Microengineering, vol. 15 (2), pp. S152 – S164, 2005. [21] V. Kolchuzhin, J. E. Mehner, T. Gessner, and W. Doetzel, “Parametric finite element analysis for reduced order modeling of MEMS,” International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Micro-Systems, (20-23 April), pp. 1 – 6, 2008. [22] S. E. Lyshevski, “Modeling and control of MEMS with high speed synchronous micromotors and controllers/drivers-on-VLSI-chip ICs,” Energy Conversion and Management, vol. 44, pp. 667 – 679, 2003. [23] S. De and N. Aluru, “Physical and reduced-order dynamic analysis of MEMS,” International Conference on Computer Aided Design, 9-13 Nov. 2003, pp. 270 – 273, 2003. [24] G. Wu, A. R. Mirza, S. K. Gamage, L. Ukrainczyk, N. Shashidhar, G. Wruck, and M. Ruda, “Design and use of compact lensed fibers for low cost packaging of optical MEMS components,” J. of Micromechanics and Microengineering, vol. 14, pp. 1367 – 1375, 2004. [25] B. Zhang and M. T. E. Kahn, “Overview and improving fiber optic gyroscope based on MEMS/NEMS fabrication,” Proceedings of the International MEMS Conference, vol. 34, pp. 148 – 154, 2006. [26] T. Bechtold, E. B. Rudnyi, and J. G. Korvink, “Automatic order reduction of thermo-electric models for mems: Arnoldi versus guyan,” Fourth International Conference onAdvanced Semiconductor Devices and Microsystems, (14-16 October), vol. 3, pp. 333 – 336, 2002. [27] J. S. Han, E. B. Rudnyi, and J. G. Korvnik, “Efficient optimization of transient dynamic problems in MEMS devices using model order reduction,” Journal of Dynamic Systems, Measurement and Control, vol. 15, pp. 822 – 832, 2005. [28] C. Chen and C. Lee, “Design and modeling for comb drive actuator with enlarged static displacement,” Sensors and Actuators A, vol. 115, pp. 530 – 539, 2004. [29] W. Zhang, G. Meng, and H. Li, “Adaptive vibration control of micro- cantilever beam with piezoelectric actuator in MEMS,” International Journal of Advanced Manufacturing Technology, vol. 28, pp. 321 – 327, 2006. [30] J. Chen, S.-M. Kang, J. Zou, C. Liu, and J. E. Schutt-Aine, “Reducedorder modeling of weakly nonlinear MEMS devices with taylor-series expansion and arnoldi approach,” IEEE Journal of Microelectromechanical Systems, vol. 13, no. 3, pp. 441 – 451, June 2004. [31] X. Lin, Z. Chen, and J. Ying, “Macromodeling of the electrostatically actuated rectangle plate based on modal projection,” International Conference on Mechatronics and Automation, (5-8 August), pp. 2741 – 2746, 2007. [32] G. K. Fedder, “Issues in MEMS macromodeling,” 2003 International Workshop on Behavioral Modeling and Simulation, (7-8 October), pp. 64 – 69, 2003. [33] J. Mehner, J. Wibbeler, and F. Bennini, “Ansys multiphysics capabilities for mems modeling and simulation -part 3: Exporting macromodels for circuit and system simulation tools.” ANSYS Solutions, www.ansys.com, 2008. [34] G. Lorenz, A. Morris, and I. Lakkis, “A top-down design flow for moems,” IEEE International Conference on Micro-Electo-Mechanical Systems, Jan. 20-24, 2002, Las Vegas, NV, USA, pp. 204 – 209, 2002. [35] ——, “A top-down design flow for MOEMS,” SPIE -4408, Design, Test, Integration, and Packaging of MEMS/MOEMS, (25-27 April) Cannes, France, pp. 126 – 137, 2001. [36] G. K. Fedder and Q. Jing, “A hierarchical circuit-level design methodology for microelectromechanical systems,” IEEE Transactions on Circuits and Systems-II, vol. 46, no. 10, pp. 1309 – 1315, September 1999. [37] W. Xue, J. Wang, and T. Cui, “Modeling and design of polymer- based tunneling accelerometers by ANSYS/MATLAB,” IEEE/ASME Transactions on Mechatronics, vol. 10, no. 2, pp. 468 – 472, April 2005. [38] Y. Xu and N. Aluru, “Multiscale electrostatic analysis of silicon nems via heterogeneous quantum models,” Physical Review B,, vol. 77 (7), p. 075313, 2008. [39] B. Borovic, C. Hong, X. M. Zhang, A. Q. Liu, and F. L. Lewis, “Open vs. closed-loop control of the MEMS electrostatic comb drive,” Proceesings of the 13th Mediterranean Conference on Control and Automation, 2005. [40] F. Shi, P. Ramesh, and S. Mukherjee, “Dynamic analysis of microelectro- mechanical systems,” International Journal for Numerical Methods in Engineering, vol. 39, no. 24, pp. 4119 – 4139, December 1998. [41] Y.-J. Yang, S.-Y. Cheng, and K.-Y. Shen, “Macromodeling of coupled- domain MEMS devices with electrostatic and electrothermal effects,” J. of Micromechanics and Microengineering, vol. 04, pp. 1190 – 1196, 2004. [42] J. Xu, W. Yuan, H. Chang, X. Lu, and Y. Yu, “Hybrid macromodels for modeling and simulation of a z-axis micro accelerometer,” 3rd IEEE International Conference on Nano/Micro Engineered and Molecular Systems, (6-9 January), pp. 357 – 361, 2008. [43] N. M. Ghoniem, E. N. Busso, N. Kioussis, and H. Huang, “Multiscale modelling of nanomechanics and micromechanics: an overview,” Philosophical Magazine, vol. 83, pp. 3475 – 3528, 2003. [44] C.-H. Ke, H. D. Espinosa, and N. Pugno, “Numerical analysis of nanotube based NEMS devices. part ii: Role of finite kinematics, stretching and charge concentrations.” Journal of Applied Mechanics, vol. 72, pp. 726 – 731, 2005. [45] W. W. Liou, F. Liu, and Y. Fang, “Navier stokes and dsmc simulations of forced chaotic microflows,” 33rd AIAA Fluid Dynamics Conference and Exhibit, pp. AIAA 2003–3583, 2003. [46] B.-Y. Cao, J. Sun, M. Chen, and Z.-Y. Guo, “Molecular momentum transport at fluid-solid interfaces in MEMS/NEMS: A review,” International Journal of Molecular Science, vol. 10, 2009. [47] A. Pantazi, A. Sebastian, G. Cherubini, M. Lantz, H. Posidis, H. Rothuizen, and Eleftheriou, “Control of mems-based scanning probe data-storage devices,” IEEE Transactions on Control Systems Technology, vol. 15, no. 5, pp. 824 – 841, September 2007. [48] Desquenes, S. V. Rotkin, and N. R. Alaru, Nanotechnology, vol. 13, no. 6, pp. 120 – 901, December 2002. [49] C. Ke, N. Pugno, B. Peng, and H. Espinosa, “Experiments and modeling of carbon nanotube based nems device,” Journal of the Mechanics and Physics of Solids, vol. 53, pp. 1314 – 1333, 2005. [50] J. W. L. Zhou, H.-Y. Chan, T. K. H. To, K. W. C. Lai, and W. J. Li, “Polymer MEMS actuators for underwater micromanipulation,” IEEE/ASME Transactions on Mechatronics, vol. 9, no. 2, pp. 334 – 342, June 2004. [51] N. Kacem, S. Hentz, D. Pinto, B. Reig, and V. Nguyen, “Nonlinear dynamics of nanomechanical beam resonators: improving the performance of nems-based sensors,” Nanotechnology, vol. 20, p. 275501, 2009. [52] C.-H. Ke and H. D. Espinosa, “Numerical analysis of nanotube based nems devices. part i: Electrostatic charge distribution on multiwalled nanotubes,” Journal of Applied Mechanics, vol. 72, pp. 721 – 725, 2005. [53] Z. Tang, Y. Xu, G. Li, and N. R. Aluru, “Physical models for coupled electromechanical analysis of silicon nanoelectromechanical systems,” Journal of Applied Physics, vol. 97, no. 11, pp. 114 304.1 – 114 304.13, November 2005. [54] J. B. Muldavin and G. M. Rebeiz, “Nonlinear electro-mechanical modeling of MEMS switches,” IEEE MTT-S International Microwave Symposium Digest, (20-25 May), vol. 3, pp. 2119 – 2122, 2001. [55] A. Lazaro, D. Girbau, L. Pradell, and A. Nebot, “Non linear actuation model for lateral electrostatically-actuated dc-contact rf MEMS series switches,” Spanish Conference on Electron Devices, Jan. 31-Feb. 2, pp. 181 – 184, 2007. [56] A. Boni, G. Fontana, and F. Pianegiani, “Time-domain modeling and characterization of capacitive mems switches,” IEEE Conference on Instrumentation and Measurement Technology Conference, 16-19 May 2005, vol. 3, pp. 1808 – 1811, 2005. [57] J. A. Pelesko, “Mathematical modeling of electrostatics mems with tailored dielectric properties,” SIAM Journal of Applied Mathematics, vol. 62, no. 3, pp. 888 – 908, February 2002. [58] Q. Qi, “A computationally efficient model for analyzing an electrostatically driven MEMS device embedded in a dielectric fluid,” 7th International Conference on Electronic Packaging Technology, vol. 14, no. 5, pp. 1 – 9, October 2006. [59] G. Schrag and G. Wachutka, “Physically-based modeling of squeeze film damping by mixed level simulation”,,” Sensors and Actuators A., pp. 64 – 69, 2001. [60] E. S. Hung and S. D. Senturia, “Dynamical models for microelectromechanical systems from a few finite-element simulation runs,” IEEE/ASME Journal of Microelectromechanical Systems, vol. 8, no. 2, pp. 280 – 289, September 1999. [61] C. Ke and H. D. Espinosa, “Nanoelectromechanical systems and modeling,” Handbook of Theoretical and Computational Nanotechnology, Vol.1, Chapter 121, pp. 1 – 38, 2005. [62] H. Reid and J. White, “Boundary-element modeling of nanoscale device electrostatics,” In Sensors, Proceedings of the IEEE, vol. 3, pp. 1403 – 1406, 2009. [63] V. Kolchuzhin, W. Doetzel, and J. Mehner, “A derivatives-based method for parameterization of MEMS reduced order models,” International SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Micro-Systems, 20-23 April 2008, pp. 1 – 5, 2005. [64] M. I. Younis, E. M. Abdel-Rahman, and A. Nayfeh, “A reduced- order model for electrically actuated microbeam-based MEMS,” IEEE Journal of Microelectromechanical Systems, vol. 12, no. 5, pp. 672 – 680, October 2003. [65] E. S. Piekos and K. S. Breuer, “Numerical modeling of micromechanical devices using the direct simulation monte carlo method,” Journal of fluids engineering, vol. 118, pp. 464 – 469, 1996. [66] J. Xu, W. Yuan, H. Chang, X. Lu, and Y. Yu, “Octree-search kinetic monte carlo algorithm for the simulation of complex 3d mems structures,” Micro Electro Mechanical Systems, 2008. MEMS 2008. IEEE 21st International Conference on, pp. 323 – 326, 2008. [67] H. Somada, K. Hirahara, S. Akita, and Y. Nakayama, “A molecular linear motor consisting of carbon nanotubes,” Nano Letters, vol. 9, no. 1, pp. 62 – 65, 2009. [68] L. Dong, X. Tao, M. Hamdi, L. Zhang, X. Zhang, A. Ferreira, and B. J. Nelson, “Nanotube boiler: Attogram copper evaporator driven by electric current, joule heating, charge, and ionization,” IEEE Transactions on Nanotechnology, vol. 8, pp. 565 – 568, 2009. [69] S. P. et al., “Molecular dynamic study on-size dependent elastic properties of the silicon nanocantilevers,” Thin Solid Films, vol. 492, pp. 285 – 289, 2005. [70] S. Zhang, W. Kim, and R. S. Ruoff, “Atomistic simulations of double- walled carbon nanotubes as rotational bearings,” Nano Letters, vol. 4, pp. 293 – 297, 2004. [71] R. E. Rudd, “Coarse-grained molecular dynamics and multiscale modeling of NEMS resonators,” Technical Proceedings of the International Conference on Computational Nanoscience and Nanotechnology, 2002. [72] K. Esfarjani and G. A. Mansoori, “Statistical mechanical modeling and its application to nanosystems,” Handbook of Theoretical and Computational Nanotechnology, vol. 1, pp. 1 – 45, 2005. [73] S. J. Tans, M. H. Devoret, H. Dal, A. Thess, R. Smalley, L. Geerligs, and C. Dekker, “Single-wall carbon nanotube as quantum wires,” Nature, vol. 386, pp. 474 – 477, 1997. [74] D. H. S. Maithripala, B. D. Kawade, I. Wickramasinghe, J. M. Berg, and W. P. Dayawansa, “A passivity-based controller for an electrostatic MEMS model in the presence of parasitics,” International Conference on Industrial and Information Systems, (9-11 August), pp. 267 – 272, 2007. [75] S. Mariani, A. Ghisi, A. Corigliano, and S. Zerbini, “Modeling impact- induced failure of polysilicon nems: A multi-scale approach,” Sensors, vol. 9, pp. 556 – 567;, 2009. [76] N. Aluru, “Hierarchical physical models for analysis of electrostatic nanoelectromechanical systems (nems),” Automatica, vol. http://nanohub.org/resources/850, 2006. [77] M. Bernashi, S. Melchionna, S. Succi, M. Fyta, and E. Kaxiras, “Quantized current blockade and hydrodynamic correlations in biopolymer translocation through nanopores: Evidence from multiscale simulations,” vol. 8, no. 4, pp. 1115 – 1119, 2008. [78] M. Hamdi and A. Ferreira, “Dna nanorobotics,” Microelectronics Journal, vol. 39, pp. 1051 – 1059, 2008. [79] ——, “Computational design and multiscale modeling of a nanoactuator using dna actuation,” Nanotechnology, vol. 20, p. 485501, 2009. [80] T. Karakasidis and C. Charitidis, “Multiscale modeling in nanomaterials science,” Materials Science and Engineering C, vol. 27, pp. 1082 – 1089, 2007. [81] A. C. T. et al., “Materials integrity in microsystems: a framework for a petascale predictive-science-based multiscale modeling and simulation system,” Comput. Mechanics, vol. 42, pp. 485 – 510, 2008. [82] M. Odegard, T. S. Gates, K. E. Wise, C. Park, and E. J. Siochi, “Effect of nanotube functionalisation on the elastic properties of polyethylene nanotube composites,” AIAA Journal, vol. 43, no. 8, pp. 1828 – 1835, August 2005. [83] H. Kubo, M. Ciappa, T. Masunaga, and W. Fichtner, “Multiscale simulation of aluminium thin films for the design of hihly-reliable mems devices,” Microelectronics Reliability, vol. 49, pp. 1278 – 1282, 2009. [84] H. Adalsteinsson, B. J. Debusschere, K. R. Long, and H. N. Najm, “Components for atomistic-to-continuum multiscale modeling of flow in micro-and nanofluidic systems,” Sci. Program., vol. 16, no. 4, pp. 297 – 313, 2008. [85] G. Lu and E. Kaxiras, “Overview of multiscale simulations of materials,” Handbook of Theoretical and Computational Nanotechnology, vol. 1, pp. 1 – 33, 2005. -PART C: APPLICATIONS AND REVIEWS [86] G. Li and N. R. Aluru, “Hybrid techniques for electrostatic analysis of nanoelectromechanical systems,” Journal of Applied Physics, vol. 96, pp. 2221 – 2231, 2004. [87] A. Lombo-Carrasquilla and G. Becerra-Forigua, “Integrated modeling methodology for nanoscale electronic devices,” 3rd IEEE International Conference on Nano/Micro Engineered and Molecular Systems, 6-9 Jan. 2008, pp. 357 – 361, 2008. [88] D. W. Wood and L. Bijnens, “Multiphysics modeling of axispherical MEMS,” 7th AFRICON Conference in Africa, vol. 2, pp. 1017 – 1022, 2004. [89] A. Ferreira, J. Agnus, N. Chaillet, and J.-M. Breguet, “A smart microrobot on chip: Design, identification and control,” IEEE/ASME Transactions on Mechatronics, vol. 9, no. 3, pp. 508 – 519, September 2004. [90] R. Oboe, R. Antonello, E. Lasalandra, G. S. Durante, and L. Prandi, “Control of a z-axis MEMS vibrational gyroscope,” IEEE/ASME Transactions on Mechatronics, vol. 10, no. 4, pp. 364 – 370, August 2005. [91] B. Borovic, F. L. Lewis, D. Agonafer, E. S. Kolesar, M. M. Hossain, and D. O. Popa, “Method for determining a dynamical state-space model for control of thermal MEMS devices,” IEEE Journal of Microelectromechanical Systems, vol. 14, no. 5, pp. 961 – 970, October 2005. [92] Y. C. Lee, P. B. Amir, J. A. Chiou, and S. Chen, “Packaging for microelectromechanical and nanoelectromechanical systems,” IEEE Transactions on Advanced Packaging, vol. 26, no. 3, pp. 217 – 226, August 2002. [93] M. Lishchynska, N. Cordero, and O. Slattery, “State of the art in prediction of mechanical behaviour of microsystems [mems],” 5th International Conference on Thermal and Mechanical Simulation and Experiments in Microelectronics and Microsystems, pp. 287 – 294, 2004. [94] J. Redmond, D. Reedy, M. Heinstein, M. de Boer, J. Knapp, E. Piekos, C. Wong, and L. Holm, “Microscale modeling and simulation,” SANDIA Report, SAND2001-3675, December 2001, pp. 287 – 294, December 2001. [95] M. Innocent, P. Wambacq, S. Donnay, H. A. C. Tilmans, W. Sansen, and H. D. Man, “An analytic volterra-series-based model for a MEMS variable capacitor,” IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 22, no. 2, pp. 124 – 131, February 2003. [96] V. Litovski, M. Andrejevic, and M. Zwolinski, “Ann based modeling, testing and diagnosis of MEMS: Capacitive pressure transducer example,” 7th Seminar on Neural Network Applications in Electrical Engineering, 23-25 Sept. 2004, pp. 183 – 188, 2004. [97] L. G. Bleris and M. V. Kothare, “Real-time implementation of model predictive control,” American Control Conference, 8-10 June 2005, vol. 6, pp. 4166 – 4171, 2005. [98] M. Bazu, C. Tibeica, L. Galateanu, and V. E. Ilian, “Fuzzy-logic reliability predictions in microtechnologies,” International Conference on Computational Intelligence for Modelling, Control and Automation and International Conference on Intelligent Agents, Web Technologies and Internet Commerce, vol. 1, pp. 89 – 93, 2005. [99] P. Singh, J. Rajagopalan, R. LaFollette, C. Fennie, and D. E. Reisner, “Fuzzy logic-based microbattery controller for mems applications,” 35th Intersociety Energy Conversion Engineering Conference and Exhibit, vol. 2, pp. 747 – 751, 2000. [100] B. Borovic, A. Q. Liu, D. Popa, H. Cai, and F. L. Lewis, “Openloop versus closed-loop control of MEMS devices: choices and issues,” IEEE Journal of Microelectromechanical Systems, vol. 15, pp. 1917 – 1924, 2005. [101] W. Tang, T. Nguyen, and R. Howe, “Laterally driven polysilicon resonant microstructures,” Sensors and Actuators A, vol. 20, pp. 25 – 32, 1989. [102] W. Hofman, C. Lee, and N. MacDonald, “Monolithic three-dimensional single-crystal silicon microelectromechanical systems,” Sensors and Materials, vol. 10, pp. 337 – 350, 1998. [103] K. M. Morzinski, K. B. W. Harpsie, D. T. Gavel, and S. M. Ammons, “The open-loop control of MEMS : Modeling and experimental results,” Proceedings of SPIE, vol. 6467, pp. 64 670G.1 – 64 670G.10, 2007. [104] S. Huang, J. Tsai, D. Hah, H. Toshiyoshi, and M. C. Wu, “Open-loop operation of MEMS WDM routers with analog micromirror array,” Proceedings of the IEEE/LEOS International Conference on Optical MEMs, pp. 179 – 180, 2002. [105] J. Tsai, S. Huang, D. Hah, H. Toshiyoshi, and M. Wu, “Open-loop operation of MEMS-based 1N wavelength-selective switch with long SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS term stability and repeatability,” IEEE Photonics Technology Letters, vol. 16 (4), pp. 1041 – 1043, 2004. [106] B.-W. Yoo, J.-H. Park, Y.-H. Jang, and Y.-K. Kim, “A low-drift, open- loop controlled, single crystalline silicon micromirror with floating field-limiting shields,” Journal of Micromechanics and Microengineering, vol. 18, p. 035031(8pp), 2008. [107] C. R. Vogel and Q. Yang, “Modeling, simulation, and open-loop control of a continuous facesheet mems deformable mirror.” Journal of the Optical Society of America A: Optics, Image Science and Vision, vol. 23 (5), pp. 1074 – 1081, 2006. [108] B. D. Jensen, S. Mutlu, S. Miller, K. Kurabayashi, and J. J. Allen, “Shaped comb fingers for tailored electromechanical restoring force,” IEEE Journal of Microelectromechanical Systems, vol. 12, pp. 373 – 383, 2003. [109] H. J. Kordlar and G. Rezazadeh, “Modeling open-loop MEMS tunneling accelerometer based on circular plate,” Sensors & Transducers Journal, vol. 78(4), pp. 1083 – 1092, 2007. [110] D. Keymeulen, M. I. Ferguson, B. Oks, C. Peay, R. Terrile, Y. Cheng, D. Kim, E. MacDonald, and D. Foor, “Hardware platforms for MEMS gyroscope tuning based on evolutionary computation using open-loop and closed -loop frequency response.” Proceedings of the International Conference on Evolvable Systems, Barcelona, Spain (September 1214), 2005. [111] M. F. Daqaq, C. K. Reddy, and A. H. Nayfeh, “Input-shaping control of nonlinear MEMS,” Nonlinear Dynamics, vol. 54 (1-2), pp. 167 – 179, 2008. [112] D. O. Popa, B. H. Kang, J. T. Wen, H. E. Stephanou, G. Skidmore, and A. Geisberger, “Dynamic modeling and input shaping of thermal bimorph actuators,” IEEE International Conference on Robotics Automation (Taipei, Taiwan, 1419 Sept.), 2003. [113] J. B. Stewart, A. Diouf, Y. Zhou, and T. G. Bifano, “Open-loop control of a MEMS deformable mirror for large-amplitude wavefront control.” Journal of the Optical Society of America A: Optics, Image Science and Vision, vol. 24 (12), pp. 3827 – 3833, 2007. [114] C. Acar, S. Eler, and A. M. Shkel, “Concept, implementation, and control of wide bandwidth MEMS gyroscopes,” Proceedings of the American Control Conference (June 25 -27), pp. 1229 – 1234, 2001. [115] “System and method for canceling disturbance mems devices,” United States Patent 6909819, June 21, 2005. [116] B. Borovic, C. Hong, A. Q. Liu, L. Xie, and F. L. Lewis, “Control of a MEMS optical switch,” Proceedings of the 43rd IEEE Conference on Decision and Control (December 14 -17), vol. 5, pp. 3039 – 3044, 2004. [117] “MEMS device control with filtered voltage signal shaping,” United States Patent 7428353, September 23, 2008. [118] V. Milanovic and K. Castelino, “Sub -100 µs settling time and low voltage operation for gimbal -less two -axis scanners,” Proceedings of the IEEE/LEOS International Conference on Optical MEMS, 2004. [119] T. Fukushige, S. Hata, and A. Shimokhobe, “A new driving method for electrostatic MEMS actuators to prevent sticking,” Proceedings of the 4th euspen International Conference ( Glasgow, Scotland, UK, May June), pp. 1 – 2, 2004. [120] S. Jordan and E. Lawrence, “Vibration nullification of MEMS devices using input shaping,” Proceedings of the SPIE Conference on Damping and Isolation (March 3 -5), vol. 5052, pp. 326 – 334, 2003. [121] S. E. Lyshevski, MEMS and NEMS: structures, devices, and systems. CRC Press, Boca Raton, FL, 2002. [122] W.-T. Sung, S. Sung, J.-Y. Lee, T. Kang, Y. J. Lee, and J. G. Lee, “Development of a lateral velocity-controlled MEMS vibratory gyroscope and its performance test,” IEEE Journal of Microelectromechanical Systems, vol. 18, p. 055028(13pp), 2008. [123] H. Cai, J. H. Wu, J. L. Zhang, X. M. Wang, Y. X. Lu, and C. Liu, “Optical MEMS switch control and packaging,” Proceedings of the 5TH Conference on Electronics Packaging Technology (December 10 -12), pp. 291 – 293, 2003. [124] M. Vagiaa, G. Nikolakopoulos, and A. Tzesa, “Design of a robust PID- control switching scheme for an electrostatic micro-actuator,” Control Engineering Practice, vol. 16 (11), pp. 1321 – 1328, 2008. [125] A. Koide, K. Sato, S. Suzuki, and M. Miki, “A multistep anisotropic etching process for producing 3-d accelerometers,” In Technical Digest of the 11th Sensor Symposium, 1992. [126] K. Ikeda, H. Kuwayama, T. Kobayashi, T. Watanabe, T. Nishikawa, T. Yoshida, and K. Harada, “Three-dimensional micromachining of silicon pressure sensor integrating resonant strain gauge on diaphragm,” Sensors and Actuators A, pp. A21 – A23: 1007 – 1009, 1990. -PART C: APPLICATIONS AND REVIEWS [127] ——, “Silicon pressure sensor integrates resonant strain gauge on diaphragm,” Sensors and Actuators A, pp. A21 – A23: 146 – 150, 1990. [128] J. Bryzek, E. Abbott, A. Flannery, D. Cagle, and J.Maitan, “Control issues for MEMS,” Proceedings of the 42nd International Conference Decision and Control, pp. 3039 – 3047, 2003. [129] “MEMS device with feedback control,” United States Patent 7061660, June 13, 2006. [130] A. Shkel, R. Horowitz, A. Seshia, S. Park, and R. Howe, “Dynamics and control of micromachined gyroscopes,” Proceedings of the American Control Conference (June 2 -4), vol. 3, pp. 2119 – 2124, 1999. [131] Y. Chen, R. M’Closkey, T. Tran, and B. Blaes, “A control and signal processing integrated circuit for the jpl-boeing micromachined gyroscopes,” IEEE Trans. Control Systems Technology, vol. 13 (2), pp. 286 – 300, 2005. [132] “Integrated low power digital gyro control electronics,” United States Patent 6915215, July 5, 2005. [133] C. C. Painter and A. M. Shkel, “Active structural error suppression in MEMS vibratory rate integrating gyroscopes,” IEEE Sensors Journal, vol. 3 (5), pp. 595 – 606, 2003. [134] H. Rodjegard, D. Sandstrom, P. Pelin, M. Carlsson, M. Bohman, N. Hedenstierna, G. I. Andersson, and A. B. Imego, “A novel architecture for digital control of MEMS gyros,” In Sensors, Proceedings of the IEEE, vol. 3, pp. 1403 – 1406, 2004. [135] M. Looney, “Optimizing MEMS gyroscope performance with digital control,” Analog Devices Application Note (AN-942). [136] S. Park, R. Horowitz, and C. woo Tan, “Adaptive control for MEMS gyroscopes,” California PATH Research Report, vol. UCB-ITS-PRR2002- 11, 2002. [137] S. Park and R. Horowitz, “Adaptive control for the conventional mode of operation of MEMS gyroscopes,” IEEE Journal of Microelectromechanical Systems, vol. 12 (1), pp. 101 – 108, 2003. [138] C. Batur, T. Sreeramreddy, and Q. Khasawneh, “Sliding mode control of a simulated MEMS gyroscope,” Proceedings of the American Control Conference (8 -10 June), vol. 6, pp. 4160 – 4165, 2005. [139] J. Fei and C. Batur, “Robust adaptive control for a MEMS vibratory gyroscope,” International Journal of Advanced Manufacturing Technology (July 4), vol. 42, no. 2 –3, pp. 293 – 300, 2008. [140] Q. Zheng, L. Dong, and Z. Gao, “A novel control system design for vibrational MEMS gyroscopes,” Sensors and Actuators A, vol. 78, pp. 1073 – 1082, 2007. [141] R. P. Leland, “Adaptive control of a MEMS gyroscope using lyapunov methods,” IEEE Trans. of Control Systems Technology, vol. 14 (2), pp. 278 – 283, 2006. [142] S. Park and R. Horowitz, “Discrete time adaptive control for a MEMS gyroscope,” International Journal of Adaptive Control and Signal Processing, vol. 19 (6), pp. 485 – 503, 2005. [143] M. S.-C. Lu and G. K. Fedder, “Position control of parallel-plate microactuators for probe-based data storage,” IEEE Journal of Micro- electromechanical Systems, vol. 13(5), pp. 759 – 769, 2004. [144] X. Wu, T. Ren, and L. Liu, “Active damping of a piezoelectric MEMS acoustic sensor,” Integrated Ferroelectrics, vol. 80, pp. 317 – 329, 2006. [145] A. Izadian, L. Hornak, and P. Famouri, “Adaptive control of MEMS devices,” Proceedings of the Conference for Intelligent Systems and Control (August 14 -16), 2006. [146] P. Chu, I. Brener, P. Chuan, L. Shi-Sheng, J. I. Dadap, P. Sangtae, K. Bergman, N. H. Bonadeo, T. Chau, C. Ming, R. Doran, R. Gibson, R. Harel, J. J. Johnson, C. D. Lee, D. R. Peale, T. Bo, D. T. K. Tong, T. Ming-Ju, W. Qi, W. Zhong, E. L. Goldstein, L. Y. Lin, and J. A. Walker., “Design and nonlinear servo control of MEMS mirrors and their performance in a large port-count optical switch,” IEEE Journal of Microelectromechanical Systems, vol. 14 (2), pp. 261 – 273, 2005. [147] J. M. Dawson, J. Chen, K. S. Brown, P. Famouri, and L. A. Hornak, “Through-wafer interrogation of microstructure motion for MEMS feedback control,” SPIE Proceedings for the Conference on Miniaturized systems with micro-optics and MEMS : ( Santa Clara CA, 20-22 September), vol. 3878, pp. 281 – 292, 1999. [148] T. Seki, M. Murakami, J. Yamaguchi, and K. Oda, “High speed mirror control technique for 3D-MEMS optical switch,” IEICE Transactions on Communications (Japanese Edition), vol. 189-B, pp. 1315 – 1317, 2006. [149] P. Song-Hee, A. Aina, T. Denison, and K. Lundberg, “Feedback control for a MEMS-based high-performance operational amplifier,” Proceedings of the American Control Conference (30 June -2 July), vol. 1 (30), pp. 380 – 385, 2004. SUBMITTED TO IEEE TRANSACTIONS ON SYSTEMS, MAN AND CYBERNETICS [150] M. Afridi, A. Hefner, D. Berning, C. Ellenwood, A. Varma, B. Jacob, and S. Semancik, “MEMS-based embedded sensor virtual components for system-on-a-chip,” Solid-State Electronics, vol. 48, pp. 1777 – 1781, 2004. [151] M. Behnam, G. V. Kaigala, M. Khorasani, P. Marshall, C. J. Backhouse, and D. G. Elliott, “An integrated CMOS high voltage supply for labon- a-chip systems,” Lab On A Chip, vol. 8, pp. 1524 – 1529, 2008. [152] C. D. Garca, Y. Liu, P. Anderson, and C. S. Henry, “Versatile 3-channel high-voltage power supply for microchip capillary electrophoresis,” Lab On A Chip, vol. 3, pp. 324 – 328, 2003. [153] T. Fujimori, Y. Hanaoka, K. Fujisaki, N. Yokoyama, and H. Fukuda, “On-chip MEMS capacitive pressure sensor fabricated using standard CMOS back-end-of-line processes,” Technical Digest: Sensors Symposium, vol. 22, pp. 224 – 227, 2005. [154] O. Brand and H. Baltes, “Microsensor packaging,” Microsystems technology, vol. 7, pp. 205 – 208, 2000. [155] C. Hagleitner, A. Hierlemann, D. Lange, A. Kummer, N. Kerness, O. Brand, and H. Baltes, “Smart single-chip gas sensor microsystem,” Nature, vol. 414, pp. 293 – 296, 2001. [156] Y. Xu, C.-W. Chiu, F. Jiang, Q. Lin, and Y.-C. Tai, “A MEMS multi- sensor chip for gas flow sensing,” Sensors and Actuators A: Physical, vol. 121 (1), pp. 253 – 261, 2005. [157] M. S.-C. Lu, Z.-H. Wu, C.-E. Huang, S.-J. Hung, M.-H. Chen, and Y.-C. King, “CMOS micromachined grippers with on-chip optical detection,” IEEE Journal of Microelectromechanical Systems, vol. 17, pp. 482 – 488, 2007. [158] T. Vestad, D. W. M. Marr, and J. Oakey, “Flow control for capillary- pumped microfluidic systems,” Journal of Micromechanics and Micro- engineering, vol. 14, pp. 1503 – 1506, 2004. [159] C. J. Easley, J. M. Karlinsey, J. M. Bienvenue, L. A. Legendre, M. G. Roper, S. Feldman, M. A. Hughes, E. L. Hewlett, T. J. Merkel, J. P. Ferrance, and J. P. Landers, “A fully integrated microfluidic genetic analysis system with sample-inanswer-out capability,” Proceedings of the National Academy of Sciences, U.S.A, vol. 103, pp. 19 272 – 19 277, 2006. [160] M. G. Pollack, R. B. Fair, and A. D. Shenderov, “Electrowettingbased actuation of liquid droplet for microfluidic applications,” Applied Physics Letters, vol. 77, pp. 1725 – 1726, 2000. [161] H. Ren, R. B. Fair, and M. G. Pollack, “Automated on-chip droplet dispensing with volume control by electro-wetting actuation and capacitance metering,” Sensors and Actuators, B, vol. 98, pp. 319 – 327, 2004. [162] J. Gong and C. J. Kim, “All-electronic droplet generation on-chip with real-time feedback control for EWOD digital microfluidics,” Lab On A Chip, vol. 8, pp. 898 – 906, 2008. [163] R. Anderson, B. Kawade, K. Ragulan, D. H. S. Maithripala, J. M. Berg, R. O. Gale, and W. P. Dayawansa, “Integrated charge and position sensing for feedback control of electrostatic MEMS,” Proceedings of the Conference on Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems (March 7 -10), vol. 5765, 2005. [164] S. B. Prakash, P. Abshire, M. Urdaneta, M. Christophersen, and E. Smela, “A CMOS potentiostat for control of integrated MEMS actuators,” Proceedings of the IEEE International Symposium on Circuits and Systems (May 21 -24), pp. 5555 – 5558, 2006. [165] K. Jensen, K. Kim, and A. Zettl, “An atomic-resolution nanomechanical mass sensor.” Nature Nanotechnology, vol. 3 (9), pp. 533 – 537, 2008. -PART C: APPLICATIONS AND REVIEWS Antoine Ferreira received the M.S. and Ph.D. degrees in electrical and electronics engineering from the University of Franche-Comte, Besancon, France,in 1993 and 1996, respectively. In 1997, he PLACE was a Visiting Researcher at the ElectroTechnical PHOTO Laboratory (ETL), in Tsukuba, Japan. He is cur- HERE rently a Professor of Robotics Engineering at the Institut PRISME, Ecole Nationale Superieure dIngenieurs de Bourges (ENSI Bourges), Bourges, France. His research interests include the design, modeling, and control of micro and nanorobotic systems using active materials, micro-nanomanipulation systems, biological nanosystems, and bionanorobotics. He is an author of three books on micro and nanorobotics and more than 100 journal and conference papers and book contributions. Dr. Ferreira was the Guest Editor for different special issues in IEEE/ASME TRANSACTIONS ON MECHATRONICS in 2009, International Journal of Robotics Research in 2009, and the IEEE Nanotechnology Magazine in 2008. PLACE PHOTO HERE Sumeet S. Aphale received the Bachelors degree in electrical engineering from the University of Pune, Pune, India, in 1999, and the M.S. and Ph.D. degrees in electrical engineering from the University of Wyoming, Laramie, in May 2003 and August 2005, respectively. From 2000 to 2005, he was a member of the Hexapod Research Laboratory where he was involved in design and control of Gough-Stewart platform- type robotic manipulators. He was a member of the Australian Research Councils Center of Excellence for Complex Dynamic Systems and Control housed at the University of Newcastle, Australia from October 2005 to June 2008 and later moved to a Research Fellow appointment with the Center for Applied Dynamics Research at the University of Aberdeen, U.K. Since June 2009, he has been appointed to the New Blood Lectureship position at the School of Engineering, University of Aberdeen, Aberdeen, U.K. His current research interests include robot kinematics and control, vibration control applications as well as design and control of smart structures, nanopositioning systems and MEMS.