Systems and methods for improved control of micro-electrical-mechanical system (MEMS) electrostatic actuator

Abstract

Methods and apparatuses for sensing and adjusting lateral motion in a comb drive actuated MEMS device are provided. If lateral motion is sensed by a lateral motion sensor coupled to the comb drive actuated MEMS device, and the lateral motion is greater than a reference value, a feedback controller adjusts the lateral motion by providing a drive signal to a comb drive electrode of a comb drive actuator.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

The following drawings form part of the present specification and are included to further demonstrate certain aspects of the present invention. The figures are examples only. They do not limit the scope of the invention.

FIG. 1A shows an electrostatic comb drive actuator in the state of lateral instability.

FIG. 1B shows an electrostatic comb capacitor unit, which can be used in accordance with embodiments of this disclosure.

FIG. 2A shows a MEMS device actuated by comb-drive actuator, which can be used in accordance with embodiments of this disclosure.

FIG. 2B shows MEMS device actuated by comb-drive actuator with additional lateral comb actuator and lateral comb sensor, in accordance with embodiments of this disclosure.

FIG. 3 shows unbalanced comb-drive actuators and capacitors, which can be used in accordance with embodiments of this disclosure.

FIG. 4 shows detailed lateral comb-drive actuator and lateral comb sensor with corresponding capacitances, in accordance with embodiments of this disclosure.

FIG. 5 shows a lateral movement controller, in accordance with embodiments of this disclosure.

FIGS. 6A, 6B, 6C, and 6D show simulations of dynamic behavior of a comb-drive system, in accordance with embodiments of this disclosure.

FIGS. 7A, 7B, and 7C show simulations of dynamic behavior of a comb driver system with various applied voltages, in accordance with embodiments of this disclosure.

DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS

The disclosure and the various features and advantageous details are explained more fully with reference to the nonlimiting embodiments that are illustrated in the accompanying drawings and detailed in the following description. Descriptions of well known starting materials, processing techniques, components, and equipment are omitted so as not to unnecessarily obscure the invention in detail. It should be understood, however, that the detailed description and the specific examples, while indicating embodiments of the invention, are given by way of illustration only and not by way of limitation. Various substitutions, modifications, additions, and/or rearrangements within the spirit and/or scope of the underlying inventive concept will become apparent to those of ordinary skill in the art from this disclosure.

The present disclosure provides modeling of a comb actuator and using feedback control to manage axial and lateral motion. In one respect, a novel comb actuator structure with built-in sensors may be provided and allow for feedback control of movements. In one respect, the sensor can sense and distinguish lateral from axial motion, which provides a more effective control of comb drive actuators including control of lateral instability. Additionally, the control system may impact the design of the comb drive, mitigating the requirements on the suspension, lowering the actuation voltage and therefore, decreasing the ratio between size of the actuator and achievable deflection.

Modeling Comb Capacitor Unit (Capacitance and Force)

Both generated force and the sensitivity of the comb structure may depend on the capacitance. Therefore, a model of one basic comb capacitor unit, as shown in FIG. 1B, is developed. The capacitance as a function of x and y is given as

$\begin{array}{cc}C\left(x,y\right)={\varepsilon }_0\mathrm{TN}\left(x_0+x\right)\left(\frac{1}{a-y}+\frac{1}{b+y}\right),& \mathrm{Eq}1,\end{array}$

where N is the number of finger electrodes of the each comb drive, ε₀ is the permittivity of the vacuum, T is the thickness of the structure and x₀ is initial overlapping between the fingers.

The partial derivatives of the capacitance of Eq. 1 with respect to x and y are given as follows

$\begin{array}{cc}\frac{\partial C}{\partial x}={\varepsilon }_0\mathrm{TN}\left(\frac{1}{a-y}-\frac{1}{b+y}\right).& \mathrm{Eq}2\\ \frac{\partial C}{\partial y}={\varepsilon }_0\mathrm{TN}\left(x_0+x\right)\left(\frac{1}{{\left(a-y\right)}^2}-\frac{1}{{\left(b+y\right)}^2}\right).& \mathrm{Eq}3\end{array}$

If voltage V is applied between two electrodes, the force generated in x and y direction is given as

$\begin{array}{cc}{F}_x=\frac{1}{2}\frac{\partial C}{\partial x}V^2=\frac{1}{2}{\varepsilon }_0\mathrm{TN}\left(\frac{1}{a-y}-\frac{1}{b+y}\right)V^2.& \mathrm{Eq}4\\ F_y=\frac{1}{2}\frac{\partial C}{\partial y}V^2=\frac{1}{2}{\varepsilon }_0\mathrm{TN}\left(x_0+x\right)\left(\frac{1}{{\left(a-y\right)}^2}-\frac{1}{{\left(b+y\right)}^2}\right)V^2.& \mathrm{Eq}5\end{array}$

Every capacitance and force calculated for various comb capacitor structures in this disclosure represents an extension of this concept.

Modeling the MEMS Device with Comb Actuators

Referring to FIG. 2A, a MEMS device actuated by a comb drive actuator fabricated using a deep reactive ion etching (DRIE) technique is shown. The MEMS device may include, without limitation, a body of the device known as a shuttle, an electrostatic comb drive actuator coupled to the shuttle and configured to create a force in the x direction and torque around z axis, a suspension, and two sensors (shown as capacitive sensors) for sensing both x and y deflections. The suspension may be coupled to the shuttle at one distal end and a substrate at the other distal end, enabling limited movements and electrical connection.

It is noted that the MEMS device shown in FIG. 2A is an exemplary embodiment and one of ordinary skill in the art would recognize that the MEMS device may include other configurations, e.g., size, shape, type, etc. Additionally, the MEMS device may include more than one comb actuator (with appropriate controller relative to the dynamics of the device).

For the purpose of illustrating how feedback stabilizes lateral motion of the device, a model is derived, where the parameters of the lateral degree of freedom (DOF) model are analyzed using a combined finite element analysis (FEA) and static experimental results. Use of a model to design a controller does not affect the ability of the controller to stabilize lateral motion. However, the lateral motion significantly contributes to the simplicity and straightforwardness of the controller design.

In one embodiment, the device shown in FIG. 2A may be modeled as two second-order differential equations describing two degrees of freedom, axial displacement along x axis, and rotation around z axis. The dynamic model of the comb drive actuator can be written as

m _x {umlaut over (x)}+d _x {dot over (x)}+k _x x=F _x   Eq. 7

J _z {umlaut over (θ)}+d _θ {dot over (θ)}+k _θ θ=M _z   Eq. 8

where m_x and J_z are the effective moving mass along the x-axis and the effective moment of inertia around the z-axes, respectively. Variables d_x and d_θ describe damping, and k_x and k_θ are the stiffnesses along the x-axis and around the z-axis, respectively.

It can be seen that θ is small during the operation of the device. In the other words, a y-direction movement of the comb structure is smaller than the radius of rotation, i.e. y<<L_A. Hence, the approximation of tgθ=y/L_A≈θ may be assumed, and thus, modifying Eq. 8 to

$\begin{array}{cc}\frac{J_z}{L_A^2}\ddot{y}+\frac{d_{\theta }}{L_A^2}\stackrel{.}{y}+\frac{k_{\theta }}{L_A^2}y=-F_Y,& \mathrm{Eq}.9\end{array}$

where Eqs. 7 and 9 represent the dynamics of the comb drive actuator in a two DOF.

Forces F_x and F_y from Eqs 7 and 9 may be determined as a contribution of all forces generated by comb capacitors, such as one shown in FIG. 1B, acting on the MEMS device. For example, the structure shown in FIG. 2a has four (n=4) comb capacitances such that

$\begin{array}{cc}{F}_x=\sum _nF_{\mathrm{xCn}}=F_{\mathrm{xC}1}+F_{\mathrm{xC}2}+F_{\mathrm{xC}3}+F_{\mathrm{xC}4}& \mathrm{Eq}.10\\ F_y=\sum _nF_{\mathrm{yCn}}=F_{\mathrm{yC}1}+F_{\mathrm{yC}2}+F_{\mathrm{yC}3}+F_{\mathrm{yC}4}& \mathrm{Eq}.11\end{array}$

where n is the number of comb capacitor units. Each of eight forces from Eqs. 10 and 11 can been determined following the procedure shown in Eqs. 1 through 5.

For typical MEMS application, it may be desirable that lateral y force from Eq. 11 does not exist (e.g., F_y=0). However, due to various imperfections, F_y generally is not equal to zero. Lateral instability occurs if lateral force becomes large enough to overcome lateral stiffness.

One of the reasons for unsymmetrical lateral force is unsymmetrical geometry. To model this, a virtually symmetrical comb drive actuator is modeled as an unbalanced one by introducing Δd, as shown in FIG. 3.

In order to determine total forces F_x and F_y from Eqs. 7 and 9, the procedure from Eqs. 1-5 may be used. The capacitances C_F and C_B are composed of two comb capacitance units 10 as the one shown in FIG. 1B and can be expressed as a function position x and y as:

$\begin{array}{cc}{C}_F\left(x,y\right)=N{\varepsilon }_0T\left(x+x_0\right)\left(\frac{1}{d-\Delta d-y}+\frac{1}{d+\Delta d+y}\right)& \mathrm{Eq}.12\\ C_B\left(x,y\right)=N{\varepsilon }_0T\left(-x+x_0\right)\left(\frac{1}{d-\Delta d-y}+\frac{1}{d+\Delta d+y}\right)& \mathrm{Eq}.13\end{array}$

where C_F and C_B are capacitances generating forward and backward force capacitances with respect to the axial direction.

With the capacitances from Eqs. 12 and 13, the force in the x direction, F_x, and the force in the y direction, F_y are calculated in a similar fashion as they were in Eqs. 4 and 5 for the drive unit shown in FIG. 1B. In one embodiment, F_x and F_y are calculated as follows:

$\begin{array}{cc}{F}_x=\frac{1}{2}\frac{\partial C_F}{\partial x}V_F^2+\frac{1}{2}\frac{\partial C_B}{\partial x}V_B^2=\frac{1}{2}N{\varepsilon }_0T\left(\frac{1}{d-\Delta d-y}+\frac{1}{d+\Delta d+y}\right)\left(V_F^2-V_B^2\right)& \mathrm{Eq}.14\\ F_y=\frac{1}{2}\frac{\partial C_F}{\partial y}V_F^2+\frac{1}{2}\frac{\partial C_B}{\partial y}V_B^2=\frac{1}{2}N{\varepsilon }_0T\left[\frac{1}{{\left(d-\Delta d-y\right)}^2}-\frac{1}{{\left(d+\Delta d+y\right)}^2}\right]\hspace{1em}\left[x_0\left(V_F^2+V_B^2\right)+x\left(V_F^2-V_B^2\right)\right]& \mathrm{Eq}.15\end{array}$

where V_F and V_B are the forward- and backward-driving voltages shown in FIG. 3. Notice that if Δd=0 and y=0, the lateral force does not exist. A typical MEMS device employing a comb drive actuator is designed to work this way. Eqs. 7, 8, 14 and 15 represent the model of the MEMS device shown in FIG. 2A with an unbalanced actuator that causes lateral instability if a voltage and axial deflection condition is favorable.

Additional Comb Structures with Lateral Sensing and Actuation

FIG. 2B shows the MEMS device with comb actuator of FIG. 2A with additional comb drives and a sensor for lateral motion. In one embodiment, the structure including both the actuator and sensor may allow substantially zero force generated in x-direction. The lateral actuator may be based on a four-element actuation structure that allows the generation of positive and negative forces in the lateral direction. The lateral sensor may be based on a four-element bridge structure that allows the discrimination of positive and negative motion in the lateral direction.

The lateral comb-like structures in FIG. 2B may include both movable and fixed electrodes. Voltage V_aT applied to the actuating electrodes may cause generation of electrostatic forces and movement of the device. Motion is sensed by connecting the capacitive sensors to bridges and observing the capacitance change on the fixed electrodes.

The lateral actuators can include top and bottom comb drive structures designed to generate force in the y direction. These comb drive structures are unbalanced with unequal gaps (a and b) between the electrodes, as shown in FIG. 4. The maximum generated force per comb area occurs when the ratio between the smaller and the larger electrode gap. For example, the ratio between a and b is about 0.42, although other ratios may be suitable. The smallest gaps are defined by the minimum processing geometry, e.g., if minimum processing gap is 2.5 μm, gap is then about 2.5 μm wide and b is about 6 μm wide, although other dimensions can be used.

The lateral sensor has a similar gap geometry in order to achieve maximum sensitivity. The number of fingers, N, and the initial electrode overlapping, x₀, may be different than with the lateral actuator. Movable capacitors are connected to the bridge structure through serial capacitors. Deflection in the y-direction can be determined from the difference between voltages.

Similar to the unbalanced comb actuator on the MEMS device shown in FIG. 2A, the actuator of FIG. 2B may also be modeled using Eqs. 12 through 15. Additionally, the force generated by the lateral actuator and lateral sensor shown in FIG. 2B may further be modeled in a similar fashion. The capacitance of the two parallel top actuating capacitors, C_TP, with respect to the x- and y-directions is given as:

$\begin{array}{cc}{C}_{\mathrm{TPa}}\left(y\right)=2{\varepsilon }_0{\mathrm{TN}}_ax_{a0}\left(\frac{1}{a-y}+\frac{1}{b+y}\right)& \mathrm{Eq}.16\end{array}$

Similarly, the two parallel bottom capacitances, C_BT, are given as:

$\begin{array}{cc}{C}_{\mathrm{BTa}}\left(y\right)=2{\varepsilon }_0{\mathrm{TN}}_ax_{a0}\left(\frac{1}{a+y}+\frac{1}{b-y}\right)& \mathrm{Eq}.17\end{array}$

Both capacitances in Eqs. 17 and 18 do not depend on x. Consequently, their contribution to the force along x-axis does not exist. The unbalance coefficient Δd is omitted for the case of a lateral actuator and sensor, because it is assumed that the sensing voltage is too low to influence lateral instability, and the actuation voltage is assumed to be an issue for the controller.

The total force in the lateral direction, F_ya, may be calculated using a similar procedure as in Eqs. 14 and 15

$\begin{array}{cc}{F}_{\mathrm{ya}}={\varepsilon }_0{\mathrm{TN}}_ax_{a0}\left[\left(\frac{1}{{\left(a-y\right)}^2}-\frac{1}{{\left(b+y\right)}^2}\right)V_{\mathrm{aT}}^2+\left(\frac{1}{{\left(b-y\right)}^2}-\frac{1}{{\left(a+y\right)}^2}\right)V_{\mathrm{aB}}^2\right]& \mathrm{Eq}.18\end{array}$

Following a similar procedure, sensing comb capacitances and force generated by the capacitances are modeled as:

$\begin{array}{cc}{C}_{\mathrm{TPs}}\left(y\right)=2{\varepsilon }_0{\mathrm{TN}}_sx_{s0}\left(\frac{1}{a-y}+\frac{1}{b+y}\right)& \mathrm{Eq}.19\\ C_{\mathrm{BTs}}\left(y\right)=2{\varepsilon }_0{\mathrm{TN}}_sx_{s0}\left(\frac{1}{a+y}+\frac{1}{b-y}\right)& \mathrm{Eq}.20\\ F_{\mathrm{ys}}={\varepsilon }_0{\mathrm{TN}}_sx_{S0}\left[\left(\frac{1}{{\left(a-y\right)}^2}-\frac{1}{{\left(b+y\right)}^2}\right)V_{\mathrm{sT}}^2+\left(\frac{1}{{\left(b-y\right)}^2}-\frac{1}{{\left(a+y\right)}^2}\right)V_{\mathrm{sB}}^2\right]& \mathrm{Eq}.21\end{array}$

Now, the actuating and sensing forces in Eqs. 18 and 21 are integrated into the model in Eqs. 7, 9, 10, and 11 for the device from FIG. 2B, and the model is

$\begin{array}{cc}\frac{J_z}{L_A^2}\ddot{y}+\frac{d_{\theta }}{L_A^2}\stackrel{.}{y}+\frac{k_{\theta }}{L_A^2}y=-F_Y+\frac{L_a}{L_A}F_{\mathrm{ya}}+\frac{L_s}{L_A}F_{\mathrm{ys}}& \mathrm{Eq}.22\end{array}$

Additionally, the lateral sensor model shown in FIG. 2B may further be modeled where the lateral position sensing shown in FIG. 4 is based on the difference between capacitances C_TPs and C_BTs from Eqs. 19 and 20. As the purpose of the overall system is keeping lateral motion at or about substantially zero, capacitance changes can be linearized around y=0 where

$\begin{array}{cc}\Delta C_{\mathrm{Sense}}=C_{\mathrm{TPs}}-C_{\mathrm{BPs}}={\left(\frac{\partial C_{\mathrm{TPs}}}{\partial y}\right)}_{y=0}\Delta y-{\left(\frac{\partial C_{\mathrm{BPs}}}{\partial y}\right)}_{y=0}\Delta y& \mathrm{Eq}.23\\ {\left(\frac{\partial C_{\mathrm{TPs}}}{\partial y}\right)}_{y=0}={\left[2{\varepsilon }_0{\mathrm{TN}}_sx_{S0}\left(\frac{1}{{\left(a-y\right)}^2}-\frac{1}{{\left(b+y\right)}^2}\right)\right]}_{y=0}=2{\varepsilon }_0{\mathrm{TN}}_sx_{S0}\left(\frac{1}{a^2}-\frac{1}{b^2}\right)& \mathrm{Eq}.24\\ {\left(\frac{\partial C_{\mathrm{BTs}}}{\partial y}\right)}_{y=0}={\left[2{\varepsilon }_0{\mathrm{TN}}_sx_{S0}\left(\frac{1}{{\left(b-y\right)}^2}-\frac{1}{{\left(a+y\right)}^2}\right)\right]}_{y=0}=2{\varepsilon }_0{\mathrm{TN}}_sx_{S0}\left(\frac{1}{b^2}-\frac{1}{a^2}\right)& \mathrm{Eq}.25\end{array}$

Introducing Eqs. 24 and 25 into Eq. 23 yields

$\begin{array}{cc}\Delta C_{\mathrm{Sense}}=\frac{4{\varepsilon }_0{\mathrm{TN}}_sx_{S0}}{a^2b^2}\left(b^2-a^2\right)\Delta y=K_{C\Delta y}\Delta y& \mathrm{Eq}.26\end{array}$

Eq. 26 may be rewritten as ΔC_Sense=K_CΔyy assuming substantially small y.

A differential change in capacitance can be sensed by a properly designed voltage amplifier coupled between V_sT and V_sB shown in FIG. 4. Additionally, the same or a similar sensor bridge configuration can be used to measure ΔC_sense by measuring the charge between sT and sB nodes with a differential charge amplifier.

A sense bridge is designed so that all motion except motion in the y direction (or around the z axis) is rejected. Motion in the x direction may be cancelled by adding symmetrical structure pointing into the negative direction of x. Eqs. 19 and 20 show this embodiment clearly. Any out of plane motion affects both the top and bottom part of the sensor in the same way, therefore canceling it. Sensitivity may change only for the case when Δy (or y) is relatively large, and this typically does not happen because the controller is assumed to keep lateral motion substantially small.

Feedback Control Design

In one embodiment, the present disclosure provides a feedback approach that prevents lateral pull-in and extends the working range of the comb drive actuator in the primary x-direction DOF. The primary requirement for the controller is to keep movement in the y-direction at substantially zero. Additional shaping of the dynamics may also be desirable.

The structure of an example controller is shown in FIG. 5. In order to simplify the analysis, a proportional-integral-derivative (PID) controller is implemented for the lateral DOF. Other controllers that include an integral behavior, i.e., having the ability to force steady-state control error to be substantially small, may be used instead of or in combination with the PID controller. For example, any PI or PID controller with or without gain scheduling accounting for axial x deflection, state space controllers with integral behavior, and the like may be used.

In other embodiments, controllers with features for implementing gain scheduling for square voltages may be used. In one embodiment, it is assumed that the lateral motion Δy is measurable and available. When the lateral feedback loop is closed, the sensed value of y is compared to the reference y_d=0 and the error signal may be passed through the PID controller. Saturation-type nonlinearities distribute voltages to the two channels (1302 and 1304 of FIG. 5) leading to the top and bottom y-direction comb drive electrodes, respectively. In one embodiment, the signal may be taken through the square root functions that take care of the electrostatic force being dependent on the squared value of the voltage.

In order to keep the controller as simple as possible, the lateral degree of freedom can be expressed as:

$\begin{array}{cc}\frac{J_z}{L_A^2}\ddot{y}+\frac{d_{\theta }}{L_A^2}\stackrel{.}{y}+\frac{k_{\theta }}{L_A^2}y=\frac{L_a}{L_A}F_{\mathrm{ya}}+F_D& \mathrm{Eq}.27\end{array}$

where all lateral forces except the actuator one as a disturbance to the system, F_D are considered.

Furthermore, for y=0, Eq. 16 yields:

$\begin{array}{cc}{F}_{\mathrm{ya}}={\varepsilon }_0{\mathrm{TN}}_Ax_{A0}\left(\frac{b^2-a^2}{a^2b^2}V_{\mathrm{aT}}^2+\frac{a^2-b^2}{a^2b^2}V_{\mathrm{aB}}^2\right)& \mathrm{Eq}.28\end{array}$

Merging Eqs. 27 and 28 and defining k_ab=(b²−a²)/(ab)², yields:

$\begin{array}{cc}\frac{J_z}{L_A^2}\ddot{y}+\frac{d_{\theta }}{L_A^2}\stackrel{.}{y}+\frac{k_{\theta }}{L_A^2}y=\frac{L_a}{L_A}{\varepsilon }_0{\mathrm{TN}}_ax_{a0}k_{\mathrm{ab}}\left(V_{\mathrm{aT}}^2-V_{\mathrm{aB}}^2\right)=k_L\left(V_{\mathrm{aT}}^2-V_{\mathrm{aB}}^2\right),\mathrm{where}k_L=\eta \left(L_a/L_A\right){\varepsilon }_0{\mathrm{TN}}_ax_{a0}k_{\mathrm{ab}}.& \mathrm{Eq}.29\end{array}$

As noted above, in one embodiment, the controller may be a PID type, with a gain scheduling applied to account for voltage-squared term, i.e. square root from Eq. 30, as given by

$\begin{array}{cc}u=-\mathrm{Py}-D\frac{y}{t}-I{\int }_0^ty\left(\tau \right)\tau V_{\mathrm{aT}}=\sqrt{u}u\ge 0,V_{\mathrm{aB}}=\sqrt{-u}u<0.& \mathrm{Eq}.30\end{array}$

In one embodiment, parameters of the controller (Eqs. 28 and 29) may be as P=7.5×10³ V²m⁻¹, D=5×10⁻² sV²m⁻¹, and I=7.5×10⁸ V²m⁻¹s⁻¹, although other suitable variables are may used.

Referring to FIGS. 6A-6D, simulation results for a test embodiment without and with a controller at the edge of the lateral instability is shown where an 8.96V signal was applied to the comb drive. The controller takes care of the lateral motion, not allowing the actuator to rotate around the z-axis.

A set of simulations are shown in FIGS. 7A-7C of a test embodiment where different voltages greater than or equal to the pull-in value have been applied to the actuator. It can be seen that the controller keeps the lateral motion at substantially zero. Achievable deflections may depend on how much force the lateral steering actuators can provide, and their values are several times larger than the maximum deflection in the uncontrolled case.

With the benefit of the present disclosure, those having ordinary skill in the art will comprehend that techniques claimed here may be modified and applied to a number of additional, different applications, achieving the same or a similar result. The claims cover all such modifications that fall within the scope and spirit of this disclosure.

REFERENCES

Each of the following references is incorporated by reference in its entirety:

• B. Borovic, F. L. Lewis, A. Q. Liu, E. S. Kolesar, D. O. Popa, The lateral instability problem in electrostatic comb drive actuators: modeling and feedback control, J. Micromech. Microeng. 16 (2006) 1233-1241.
• Borovic et al., In: Lateral Motion Control of Electrostatic Comb Drive: New Methods in Modeling and Sensing, 16^th IASTED Intl. Conf. Modeling and Simulation, Cancun, MX, 2005.
• Bryzek et al., In: Control Issues for MEMS, Proc. 42^nd IEEE Conf. on Decision and Control, HI, 3:9-12, 2003.
• Grade et al, J. Microelectromech. Systems, 12(3):335-343, 2003.
• Fujita, Proc. of the IEEE, 86(8):1721-1732, 1998.
• Borovic et al., In: Control of a MEMS Optical Switch, Proc. 42^nd IEEE Intl. Conf. on Decision and Control, CDC, Bahamas, 2004.
• Liu et al., Micromech. Microeng, 13:400-411, 2003.
• Legtenberg et al., Micromech. Microeng., 6:320-329, 1996.
• Elata et al., Microeletromech. Syst., 12:681-691, 2003.
• Chen and Lee, Sensors and Actuators A, 115:530-539, 2004.
• Grade, In: Large deflection, high speed, electrostatic actuators for optical switching applications, Mechanical Engineering, Stanford, 232, 1999.
• Borovic et al., J. Micromech. Microeng., 15:1917-1924, 2005.
• Senturia, In: Microsystem Design, Kluwer Academic Publ., ISBN: 0792372468, 2000.
• Elswenpoek and Wiegerink, In: Mechanical Microsensors, ISBN: 3540675825, Springer-Verlag, Berlin, 2001.
• Astrom and Wittenmark, In: Computer-Controlled Systems: Theory and Design, 294; 307; 152; Prentice-Hall, Upper Saddle River, 1997.
• Bryzek et al., IEEE, 7(2):51-59, 2004.
• Seeger and Boser, Microeletromech. Syst., 12(5):656-671, 2003.
• Franco, In: Design with operational amplifiers and analog integrated circuits, McGraw-Hill Ed., Asia, 2002.

Claims

1. A method comprising: sensing lateral motion associated with a comb drive actuated MEMS device;providing a value corresponding to the sensed lateral motion to a feedback controller;comparing the sensed lateral motion value with a reference value;providing a signal if the sensed lateral motion value is greater than the reference value, the signal being provided from the feedback controller to a comb drive electrode; andadjusting the lateral motion to substantially zero using the signal.

2. The method of claim 1, where the reference value is equal to substantially zero.

3. The method of claim 1, further comprising computing an electrostatic force of the signal based on the squared value of an input voltage applied to the comb drive actuator.

4. The method of claim 1, where sensing lateral motion comprises determining capacitance variance.

5. The method of claim 4; where the sensed lateral motion value comprises a capacitance value.

6. The method of claim 4, further comprising determining a plurality of comb capacitances in the comb drive actuator.

7. An apparatus, comprising: a comb drive actuated MEMS device;a plurality of lateral motion sensors coupled to the comb drive actuated MEMS device for sensing motion in a lateral direction;a feedback controller coupled to the lateral motion sensors for adjusting lateral motion in the comb drive actuated MEMS device, the feedback controller being configured to provide a drive signal to a comb drive electrode coupled to the comb drive actuated MEMS device.

8. The apparatus of claim 7, the feedback controller comprising a proportional-integral-derivative controller.

9. The apparatus of claim 7, the plurality of lateral motion sensors being configured in a bridge configuration.

10. The apparatus of claim 7, the plurality of lateral motion sensors comprising a plurality of capacitance sensors.

11. The apparatus of claim 7, the feedback controller being configured to adjust lateral motion in the comb drive actuator to substantially zero.