TWI584001B - Camera,actuator device and method of image stabilization - Google Patents

Description

Camera, actuator device, and image stabilization method

The present invention relates generally to optical devices and, more particularly, to a MEMS based image stabilization system.

The present application claims priority to U.S. Application Serial No. 13/247,906, filed on Sep. 28, 2011, and Serial No. 13/247,895, filed on Sep. 28, 2011. The entire contents of these applications are incorporated herein.

The explosive growth of cellular cameras with features such as zoom, auto focus, and high resolution has threatened to make fully automated digital cameras obsolete. However, due to the migration of these small cameras to higher megapixel densities and zoom capabilities, the resulting image quality has caused us dissatisfaction. In fact, even when consciously trying, it is practically impossible to keep a camera still by hand, because the human hand has a natural jitter in one of the peaks in the range of 7 Hz to 11 Hz. Depending on the exposure time and the angular field of view of each image pixel, this approximately 10 Hz vibration of the camera will have an ever-increasing effect on image quality. As a result, the increase in pixel density of mobile phone cameras has introduced more and more image blurring from camera shake.

Therefore, MEMS-based motion sensors for digital cameras have been developed to address image degradation caused by human hand flutter. For example, a MEMS-based gyroscope can be used to sense camera motion. In response to the sensed motion, an image stabilization system attempts to move the lens or image sensor to minimize or eliminate blurring caused by the resulting motion of the image. However, the resulting actuation system is used Conventional actuators are implemented.

Therefore, this technology requires a MEMS-based image stabilization system.

In accordance with a first aspect of the present invention, a camera is provided comprising: a plurality of electrostatic actuators; and an Optical Image Stabilization (OIS) algorithm module operative to command the plurality of actuators to respond The at least one lens is actuated by movement of the camera.

According to a second aspect of the present invention, a method of image stabilization is provided, comprising: sensing a motion of a camera; determining, based on the sensed motion, a lens actuation of one of the stabilized camera lenses; The desired lens actuation is converted to the desired tangential actuation; and the plurality of tangential actuators are used to tangentially actuate the at least one lens in response to the desired tangential actuation.

According to a third aspect of the present invention, there is provided an actuator apparatus comprising: a stage elastically supported to move in a plane; three or more actuators, each of which Coupled to an outer periphery of the stage and operable to apply a force acting in the plane and tangential to the stage when it is actuated; and an outer frame that surrounds and supports the load The table and the actuators.

The above and many other features and advantages of the novel actuator device of the present invention are obtained from the consideration of a detailed description of some exemplary embodiments of the present invention, in particular in conjunction with the drawings. One of several ways of using the actuator devices is better understood, wherein the same reference numerals are used to identify one or more of the drawings of the present invention. The same components.

Electrostatic MEMS-based lens actuation is utilized to provide an effective image stabilization system. In one embodiment, as few as three actuators can be placed around an optical component such as a lens to achieve image stabilization by utilizing tangential actuation. Referring now to the drawings, an image stabilization fixture 100 includes a central aperture 105 defined by a circular mounting stage 110 to receive one of an optical element such as a lens or group of lenses (not illustrated). Three actuators designated as the actuator 1, the actuator 2, and the actuator 3 are symmetrically disposed around the aperture 105. Each actuator actuates the stage 110 in an all-directional manner. In other words, by one of the linear displacements 120 introduced by each actuator, a vector direction tangentially enclosing one of the centers of the apertures 118 is defined. For example, the linear displacement 120 is tangential to a circle defined by the mounting stage 110.

Considering the Cartesian coordinate system defined at the center 118 of the aperture 105, the resulting tangential actuation is better understood. Stage 110 and actuators 1, 2 and 3 are located in one of the planes defined by the x and y directions. The z-direction protrudes from the plane normal at the center 118. As used herein, as indicated by direction 115, the all-direction displacement is considered to be the positive displacement of each actuator. In this respect, each actuator thus has a positive displacement and a negative displacement. As seen in Figure 2A, if the actuators 1, 2, and 3 each introduce an equal displacement and the actuators 1 and 2 are tangentially negatively displaced and the actuator 3 is tangentially displaced, the resulting cut of the stage 110 The actuation system is in the positive x direction. Conversely, if all of the actuators 1 and 2 are displaced and the actuator 3 is equally negatively displaced, the resulting tangential actuation of the stage 110 is in the negative x-direction, as shown in Figure 2C. or If the actuator 3 remains neutral, the actuator 1 is actuated negatively for a given amount, and the actuator 2 is positively actuated by the same amount, then the net actuation of the stage 110 is positive. Direction is as shown in Figure 2B. Conversely, if the actuator 3 remains neutral but the positive and negative displacements for the actuators 1 and 2 are switched as shown in Figure 2D, the net actuation of the stage 110 is in the negative y-direction. . In this manner, tangential actuation can achieve any desired amount of x and y displacement of the stage 110 within the travel limits of the actuators.

Tangential actuation can also be introduced to rotate the stage 110 about one of the z-axes. For example, if the actuators 1, 2, and 3 each introduce an equal amount of negative displacement, the net actuation of the stage 110 is one of the clockwise rotations (negative θ) in Figure 2E. Conversely, if the actuation of Figure 2E is all reversed as shown in Figure 2F such that all tangential actuations are positive, the net actuation of stage 110 is a counterclockwise z-axis rotation (positive θ). In this manner, stage 110 can be translated in the x and y planes as needed and stage 110 can be rotated in the θ direction.

The tangential displacement introduced by each of the actuators 1 through 3 can be represented in a partial coordinate system. For example, the actuator tangential direction of x 3 can be displaced in the predetermined displacement of the named _third direction to the same direction by the positive direction of FIG 1115 represents the sum of L. Similarly, the tangential displacement of actuators 1 and 2 can be represented by local linear coordinates L ₁ and L ₂ , respectively. It depends upon the effective cut from the center of each actuator 118 to the actuator to move radially of the point from the R, L ₁ from the displacement of the actuator in the dimension of 1, from the actuator in the dimension of the displacement of the 2 L ₂ and from the actuator 3 is displaced in the L dimension (size) of the entire ₃ translational and can stage 110 in the x and y dimensions of the stage 110, and one relating to rotation on θ. In this regard, a flag can be displayed as follows:

L ₃ =RSin θ +X

The above coordinate transformation assumes that the neutral position of the lens is at the origin, but can be modified accordingly if the neutral position is displaced from the origin. Using such coordinate transformations, a translation or rotation detected by one of the x, y planes of the stage 110 caused by camera shake or other unintended physical interference can be addressed by a corresponding tangential actuation.

Any suitable actuator can be used to construct actuators 1, 2 and 3 such as a comb actuator or a gap closure actuator. A biased comb actuator provides an attractive stroke characteristic, such as +/- 50 microns, and can be applied as disclosed in the U.S. Application Serial No. 12/946,670 (the '670 application) filed on Nov. 15, 2010. It is implemented as discussed in the case, and the contents of this case are incorporated herein by reference. In this embodiment, each actuator has a fixed portion 121 and a movable portion 122. In the image stabilization device 100 of FIG. 1, the fixed portion 121 is integral with an outer frame 125 and includes a plurality of fixed comb supports 112 that extend radially toward the movable portion 122. Similarly, the movable portion 122 includes a plurality of comb supports 113 that extend radially toward the fixed portion 121. The comb supports 112 and 113 alternate with each other to support a plurality of combs 114. For the sake of clarity of illustration, the comb 114 is not shown in Figure 1, but the comb 114 is shown in close-up in Figures 4A-4C.

As seen in more detail in FIG. 3, each of the actuators 1 through 3 drives the stage 110 through a corresponding flexure 106. To allow movement from the relative actuator, each scratch The curved piece 106 can have a relatively large flexibility in the radial direction while being relatively stiff in the tangential direction (corresponding to the linear displacement 120 of Figure 1). For example, flexure 106 can include a V-folded flexure having one of the longitudinal axes aligned in a tangential direction. This V-shaped flexure allows a radial deflection to be relatively stiff relative to the displacement 120. In this manner, a "pseudo-kinematic" placement of the stage 110 is achieved which accurately positions the center 118 in a stationary state, thereby achieving the desired x-y plane translation and θ rotation during image stabilization.

Fabrication of the comb 114 using one of the MEMS programs that achieve one of the actuators 1 to 3 biased deployment states can be accomplished using a linear deployment such as one discussed in the '670 application. As seen in the close-up view of Fig. 4A, the interdigitated fingers constituting each of the combs 114 can be manufactured in a full interdigitated state. In other words, the fingers of the comb 114 are initially positioned such that the associated fixed comb support 112 and movable comb support 113 are spaced apart by the length of the fingers in the comb 14. Therefore, applying a voltage difference across the comb 114 in the undeployed state of FIG. 4A will not cause linear movement of the stage 110 relative to the frame 125, and thus will not result in coupling to one of the stages of the stage 110. Any corresponding X, Y or θ movement. In order to make room for actuation, each comb 114 should be separated and deployed as shown in Figure 4B.

As illustrated in Figure 4B, in one embodiment, this deployment can be accomplished by moving the comb support 113 (and, therefore, the movable portion 122 of Figure 1) to the direction of an arrow 400. A position is deployed coplanar, parallel, and spaced apart by a selected distance from the associated fixed comb support 112 and then the movable portion 122 is secured in the deployed position to move substantially coplanarly and linearly relative to the fixed portion 121. As illustrated in Figure 4C, when When deployed, as indicated by a double arrow 405, applying and removing a suitable voltage difference across the comb 114 will cause the resiliently supported movable portion 122 to move substantially linearly and coplanarly toward one of the fixed portion 121 and away from the fixed portion 121. And, thus, causing one of the elements coupled to the stage 110 to move corresponding to X, Y, and/or θ.

There are several different methods and apparatus for deploying the moveable portion 122 to the deployment location and locking or securing it in the deployment location. For example, as seen in FIG. 3, a deployment method may involve a coplanar over-center latch 300 and a fulcrum 304 on the frame 125. The latch 300 is coupled to the frame 125 with a latch flexure 306. A coplanar deployment lever 308 is coupled to the movable portion 122 through a deployment flexure 310. The deployment lever 308 has a cam surface 312 that is configured to engage the latch 300. Additionally, the lever 308 has a notch for engaging the fulcrum 304 to rotationally move the lever relative to the fulcrum 304.

In an exemplary deployment, as shown in FIG. 3, an acceleration pulse is applied to the movable portion 122 in the direction of arrow 314 while the frame 125 remains stationary. This pulse causes the deployment lever 308 to rotate about the fulcrum 304 toward the latch 300. Rotation of the deployment lever 308 about the fulcrum 304 causes the cam surface 312 to engage the latch 300 as seen in FIG. Initially, the lever 308 is in an un deployed position 501, but begins to rotate into the intermediate position 502 such that the cam surface 312 biases the latch 300 and stretches the latch flexure 306. For the sake of clarity of illustration, most of the resected deployment flexures 310 are shown. Continued rotation of the lever 308 allows the latch flexure 306 to pull back the latch 300 to latch the lever 308 into a latched position 503. In order to generate the lever 308 to rotate and make the movable portion With a 122-pulse acceleration pulse, a small pin or another MEMS device can be inserted into a pull ring 315 (Fig. 3) and actuated accordingly. In an alternative embodiment, the movable portion 122 can be deployed using capillary action as described in commonly assigned U.S. Application Serial No. 12/946,657, filed on Nov. 15, 2010, the content of which is incorporated by reference. Incorporated herein. Similarly, alternative deployment and latching structures and methods are described in the '670 application.

This deployment and latching can cause the comb 114 to be relatively completely open as shown in Figure 4B. In this position, the comb 114 can only effectively contract relative to expansion. In this regard, both contraction and expansion may be desired to achieve both tangential and negative tangential movements discussed above with respect to actuators 1, 2, and 3. Thus, one of the preset states during image stabilization may involve applying a certain degree of voltage across the comb 114 to achieve the intermediate yoke as shown in Figure 4C. In this manner, if the comb voltage drops below the preset operating voltage level of Figure 4C, the comb 114 will expand. Conversely, if the comb voltage increases relative to the predetermined operational level, the comb 114 will contract. In this manner, both the positive and negative actuations can be applied by actuators 1, 2 and 3 as indicated by arrow 405.

The actuators may be in a "stroke start", "power down" or "parked" state prior to applying the preset voltage across the comb 114. In the "stopped" state, image stabilization is inactive but does not affect center 118. As discussed with respect to Figure 2F, one of each of the actuators 1, 2, and 3 is properly displaced to produce one of the positive rotations on θ, but no x-y plane translation is produced. Thus, this displacement at each comb 114 is sufficient to advance from the deployed but inactive state shown in Figure 4B to the preset operational state of Figure 4C. Figure 6A shows the rotation of the actuators 1, 2 and 3 from the rest state to the active optical image stabilization state. turn. As illustrated in Figure 6B, selectively applying a controlled increase or decrease in the respective operating voltages of the respective actuators 1, 2 and 3 after the comb 114 has been biased to its operating voltage will result in One of the stages 110 (and thus the center 118) discussed above in connection with Figures 2A-2F is deterministically movable. As shown in FIG. 6C, in order to save power when imaging is not performed, the actuators 1, 2, and 3 can be stopped again in their inactive state.

The deployment of the actuators 1, 2, and 3 from their rest state to their optical image stabilization can be better understood with reference to FIGS. 6D and 6E. The power-off state of the actuator device can be advantageously utilized to protect the device from impact forces and in-plane sag effects on the stage and flexure during periods of device inactivity. Thus, in some embodiments, one or more locking arms 308 (each having a locking feature 428 disposed thereon) can be coupled to the perimeter of the stage 110 and corresponding to a plurality of complementary locking features 430 can be coupled to, for example, the outer frame 125 and configured to engage one of the complementary locking features 428 on the locking arms 308 when the stage 110 is rotationally disposed in a rest or power down state. Figure 6F is a view in which the locking feature 430 is engaged with the complementary locking feature 428, while Figure 6G shows these features in a deployed state. In FIG. 6G, the dashed outline 432 is bounded in the operational state of the actuator device and the range of motion of the perimeter of one of the locking arms 308 and a locking feature 428 during an image stabilization operation, and is illustrated herein. No interference will occur between the arm 308, the locking feature 428 and the complementary locking feature 430 during operation.

A block diagram of a control system 700 using tangential actuation to control image stabilization is shown in FIG. In image stabilization, the expected motion of the camera is Distinguish between unintended jitter. For example, a user can deliberately move a camera over a 90 degree range of motion to image different subjects. If this intentional movement is not detected, the image stabilization system will have an impossible and undesirable task of rotating the lens by 90 degrees to compensate for this expected motion. One way to distinguish unintended jitter from the camera is to track the loop using one of the expected motions of a predictive camera. In an embodiment, control system 700 includes a tracking filter, such as Kalman filter 705, that predicts a current lens position based on previous measurements.

The Kalman filter 705 requires some measure of camera motion to make a prediction of why the camera is expected to move relative to unintended jitter. Thus, an inertial sensor, such as one of the MEMS-based gyroscopes 710, measures the speed of a reference point on the camera, such as the previously described aperture center 118. The x, y plane velocities, such as from the center 118 obtained from the pitch and roll measurements of the gyroscope 710, may be designated x _g and y _{g , respectively} . These inertial measurements can be supplemented by motion estimates obtained from the analysis camera image. Thus, a camera image processor 720 can also estimate the x, y plane velocity of the center 118 (which may be designated x _c and y _{c , respectively} ). The Kalman filter receives velocity estimates from gyroscope 710 and camera image processor 720 to filter the velocity estimates to thereby predict the x, y plane velocity of lens center 118. This Kalman filter prediction for the reference position velocity in the x, y plane can be named x ₀ and y _{0 , respectively} . The velocity estimates are filtered through a high pass filter 725 to remove the gyroscope drift and are integrated in the integrator 730 and multiplied by an appropriate scaling factor in the amplifier 735 to obtain a position estimate 740. In this regard, the estimate 740 represents the expected position of the lens center 118 predicted by the Kalman filter 705 in the absence of jitter. Any difference between the estimate 740 and the actual lens position is considered jitter and should be compensated by the image stabilization control system 700. It will be appreciated that embodiments of control system 700 that do not include such a predictive tracking loop may be implemented. For example, inertial measurements from gyroscope 710 may only be high pass filtered to provide a rough estimate of one of the expected camera speeds. The integration of these velocity estimates can be obtained as discussed above to obtain a position estimate 740.

To obtain the actual lens position (or equivalently, such as the location of a reference point of the center 118), each actuator is associated with a position sensor. For example, the actuator 1 may be one of L ₁ senses the displacement of the position sensor 741 associated with the earlier discussed. In this regard, position sensor 741 can sense across the comb capacitor 114 in _a displacement of the estimate made by the L. Alternatively, other types of position sensors can be used, such as a Hall sensor. Similarly, actuators 2 and 3 are associated with corresponding position sensors 742 and 743. The position sensor 742 thus senses the L ₂ displacement, while the sensor 743 senses the L ₃ displacement. The sensed displacements can then be digitized in the corresponding analog-to-digital converter 745 and the sensed displacements presented to a standard translator 750. The tangential actuations L ₁ through L _{3 can be} converted to a sensing position x _s , y _s by recalculating the previously discussed equations with θ equal to zero. Next, an adder 755 is used to determine the difference between the sensed position and the position predicted by the Kalman filter. The output from adder 755 can then be filtered in controller 760 and compensator 765 to obtain the resulting x and y coordinates of the camera that should be actuated to compensate for camera shake.

The like a translator 770 translates coordinates x and y in the equation described above cut θ to zero the coordinate L _1, L ₂ and L _3. Thus, the output from translator 770 represents the actuation of actuators 1 through 3. The reason for the Kalman filter prediction and the resulting actuation to occur at a relatively slow data rate is that a large amount of computation is required. However, the actual actuation of the actuators 1 through 3 to the desired degree of actuation can occur at a relatively high data rate. Thus, one of the separation lines 771 in FIG. 7 indicates that the digital domain of one of the control systems 700 is segmented into a relatively high data rate and a relatively low data rate. Similarly, a separation line 772 indicates that the control system 700 is partitioned into a digital domain and an analog domain.

A corresponding adder 775 can be used to determine the difference between the degree of actuation of the actuators 1, 2, and 3 and the actual actuation. Next, a corresponding controller 780 thus determines an appropriate control signal for one of its actuators. The resulting digital control signal can then be converted to an analog control signal using a digital to analog converter (DAC) 785. As is known in the art, an electrostatic comb actuator typically requires a boost voltage level such as that obtained by a charge pump. Therefore, each of the actuators 1 to 3 is driven by a corresponding driver circuit 790 in response to an analog control signal generated in the DAC 785. In this manner, control system 700 can advantageously achieve image stabilization using a gyroscope 710 that senses camera motion in the Cartesian x, y plane to use only three tangential MEMS actuators.

Image stabilization using system 700 can be implemented using a number of alternative embodiments. In this regard, the sum of the digital components and signal paths from the Kalman filter 705 through the translators 770 and 750 can be named an OIS algorithm module. The OIS algorithm module can be implemented in various integrated circuit architectures. As shown in FIG. 8, one embodiment of a camera 800 includes an OIS algorithm module 805 within a MEMS driver integrated circuit (IC) 810. Camera 800 includes a MEMS tangential actuator for image stabilization as discussed above and for self Actuator for moving focus (AF) purposes and zooming purposes. These MEMS actuators are collectively shown as a MEMS module 815. The driver IC 810 drives the MEMS module 815 with an AF command 820 from an AF driver 830 and an in-plane tangential actuation command 825 from an optical image stabilization (OIS) driver 835. The MEMS module 815 includes a position sensor such as that discussed with respect to FIG. 7 such that the driver IC 810 can receive an in-plane tangential actuator position 840.

A busbar, such as one of the I ² C bus bars 845, couples the driver IC 810 to other camera components. However, it should be understood that other bus arrangements can be utilized. In camera 800, gyroscope 710, imager 720, an image processor 850, and a microcontroller unit (MCU) 855 are all coupled to I ² C bus 845. Because the I ² C protocol is a master-slave protocol, the location of the module 805 in the driver IC 810 provides a lower latency as will be further described herein. FIG. 9 shows the resulting control loop for camera 800. The bus master can be an ISP or MCU as represented by the master module 900. The OIS algorithm module 805 is a simplified version in which the tracking filter is omitted and the expected motion of the camera is approximated by high pass filtering 910 of the pitch and roll rates from the gyroscope 710. Since the data stream on a master-slave bus is always active to active or self-active to slave, the rate of rotation from the gyroscope 710 first flows to the master module 900 and then to the driver IC 810. In this regard, the master module 900 controls both the gyroscope 710 and the driver IC 810. For the sake of clarity of illustration, only a single combined channel for one of the OIS algorithm modules 805 is shown. Thus, a translator 920 represents the translators 770 and 750 of FIG. The actual and desired lens position is translated relative to a lens neutral position 925 within the translator 920.

The resulting data traffic on bus 845 is shown in FIG. Image stabilization must capture a certain amount of current and it is therefore desirable to begin image stabilization only when a user takes a digital photo. At this point, the OIS data service begins in an initial step 1000 in which the master module 900 acts as an I ² C bus master. At this point, the gyroscope 710 can begin to take inertial measurements on camera movement and as represented by step 1005, the OIS driver 835 can command the MEMS actuator 815 to transition from a stopped state to an active state. Next, in a step 1010, the master module 900 reads the 6-bit gyro data so that the data can be written to the driver IC in a step 1015. Next, in a step 1020, the OIS algorithm module 805 can determine the appropriate amount of actuation to address the camera shake. If the user has finished taking a digital photo (as determined in step 1025), the process ends at step 1030. Otherwise, steps 1010 through 1025 are repeated. The communication time for one cycle (steps 1010 to 1020) depends on the bus cycle period and the data width. If the bus 845 can accommodate 3 bytes in each 10 μs clock cycle, the cycle time is 10 μs × 2 × 6 × 8 + the algorithm calculation time of step 1020, which is equal to 0.96 ms + algorithm calculation time.

An alternative control architecture in which the OIS algorithm module 805 is located within the ISP 850 is shown in FIG. Similar to FIG. 9, one of the ISP 850 autofocus algorithm modules 940 controls the AF driver 830 in the driver IC 810. Driver IC 810, gyroscope 710, imager 720, ISP 850, and MCU 855 all communicate using I ² C bus 845. Figure 12 shows the resulting control loop. The OIS algorithm module 805 is in turn a simplified version in which the tracking filter is omitted and the expected motion of the camera is approximated by high pass filtering 910 of the pitch and roll rates from the gyroscope 710. The ISP 850 controls both the gyroscope 710 and the driver IC 810. For the sake of clarity of illustration, only a single combined channel for one of the OIS algorithm modules 805 is shown.

The resulting data traffic for use on the bus 845 of the embodiment of Figures 11 and 12 is shown in FIG. In response to a call to an active image capture mode, the OIS data traffic begins in an initial step 1300 in which the ISP 850 acts as an I ² C bus master. Alternatively, the MCU 855 can act as a master. Simultaneously with or after step 1300, gyroscope 710 may begin to take inertial measurements on camera movement and, as represented by step 1305, OIS driver 835 may command MEMS actuator 815 to transition from a stopped state to an active state. Next, in a step 1310, the ISP 850 reads the 6-bit gyro data. Further, in a step 1315, the ISP 850 reads the current lens position from the translator 920 as data for six bytes. Next, in a step 1320, the OIS algorithm module 805 can determine the appropriate amount of actuation to address the camera shake, and in a step 1325, the ISP 850 can therefore write the driver IC with a six-byte actuation command. If the user has finished taking a digital photo (as determined in step 1330), the process ends at step 1335. Otherwise, steps 1310 through 1325 are repeated. The communication time for one cycle (steps 1310 to 1325) depends on the bus cycle period and the data width. If the bus 845 can accommodate 3 bytes in each 10 μs clock cycle, the cycle time is 10 μs × 3 × 6 × 8 + the algorithm calculation time of step 1320, which is equal to 1.44 ms + algorithm calculation time. Thus, in a master-slave busbar protocol system, positioning the OIS algorithm module 805 in the IC driver 810 as previously discussed is faster. In contrast, positioning the OIS algorithm 805 in the ISP 850 requires an extra step of data movement.

Many modifications to the materials, equipment, configurations, and methods of use of the actuator device of the present invention can be made without departing from the spirit and scope of the present invention, as will be apparent to those skilled in the art. And the scope of the present invention should not be limited to the scope of the specific embodiments, but should be accompanied by the following description, and the specific embodiments illustrated and described herein are merely examples of the present invention. The scope of the patent application and its functional equivalents are fully commensurate.

1‧‧‧Actuator

2‧‧‧Actuator

3‧‧‧Actuator

100‧‧‧Image Stabilization Accessories/Image Stabilizer

105‧‧‧Center aperture

106‧‧‧Flexing pieces

110‧‧‧Circular mounting stage

112‧‧‧Fixed comb support

113‧‧‧Removable comb support

114‧‧ comb

115‧‧‧ Direction

118‧‧‧Pore Center/Lens Center

120‧‧‧linear displacement

121‧‧‧Fixed part

122‧‧‧ movable part

125‧‧‧External framework

300‧‧‧Coplanar eccentric latch

302‧‧‧Latch

304‧‧‧ pivot

306‧‧‧Latch flexure

308‧‧‧Deploying lever/latch lever/locking arm

310‧‧‧Deploying flexures

312‧‧‧ cam surface

314‧‧‧ arrow

315‧‧‧ pull ring

400‧‧‧ arrow

405‧‧‧ arrow

428‧‧‧Locking feature

430‧‧‧Complementary locking feature

432‧‧‧dotted outline

501‧‧‧Undeployed location

502‧‧‧ intermediate position

503‧‧‧Latch position

700‧‧‧Image Stabilization Control System

705‧‧‧ Kalman filter

710‧‧‧Gyro

720‧‧‧ Camera Image Processor/Imager

725‧‧‧High-pass filter

730‧‧‧ integrator

735‧‧Amplifier

740‧‧‧Location Estimate

741‧‧‧ position sensor

742‧‧‧ position sensor

743‧‧‧ position sensor

745‧‧‧ Analog to Digital Converter (ADC)

750‧‧‧Coordinate Translator

755‧‧‧Adder

760‧‧‧ Controller

765‧‧‧ compensator

770‧‧‧Translator

771‧‧‧ separate line

772‧‧‧ separate line

780‧‧‧ controller

785‧‧‧Digital to analog converter (DAC)

790‧‧‧Drive circuit

800‧‧‧ camera

805‧‧‧Optical Image Stabilization (OIS) Algorithm Module

810‧‧‧Drive integrated circuit (IC)

815‧‧‧Micro Electro Mechanical Systems (MEMS) Modules

820‧‧‧Auto Focus (AF) command

825‧‧‧In-plane tangential actuation command

830‧‧‧Auto Focus (AF) Driver

835‧‧‧Optical Image Stabilization (OIS) Driver

840‧‧‧In-plane tangential actuator position

845‧‧‧I ² C busbar

850‧‧ Image Processor (ISP)

855‧‧‧Microcontroller Unit (MCU)

900‧‧‧Master Module

910‧‧‧High-pass filtering

920‧‧‧Translator

925‧‧‧ lens neutral position

940‧‧‧Auto Focus (AF) Algorithm Module

R‧‧‧radius distance

1 is a plan view of an exemplary image stabilization device utilizing tangential actuation; FIGS. 2A-2F illustrate the use of the exemplary image stabilization device of FIG. 1 to achieve in-plane translation and rotational movement of an optical component. Figure 3 is a perspective view of one of the actuators of the apparatus of Figure 1; Figure 4A is a partial plan view of one of the fingers of the comb of Figure 3, shown in the deployment The actuator is a finger prior to operation; FIG. 4B is a partial plan view of one of the fingers of the comb of FIG. 3 showing the fingers after the actuator has been deployed Figure 4C is a partial plan view of one of the fingers of the comb after one of the actuators of Figure 3 has been biased to an operating position; Figure 5 is a plan view of one of the actuator latches of Figure 3. , which is shown in various stages in the engagement of the actuator latch with the actuator lever; FIGS. 6A-6C illustrate the "stop" shape of the device of FIG. A vector diagram of the in-plane rotational movement of the stage between an "operating" state and a "operating" state; Figure 6D is a plan view of one of the stages in a stopped state; Figure 6E is a plan view of one of the stages in an operational state Figure 6F is a close-up view of one of the locking stage arms of Figure 6D.

Figure 6G is a close-up view of one of the unlocked stage arms of Figure 6E.

Figure 7 is a block diagram of an image stabilization system using a tangential actuator; Figure 8 is a block diagram of one embodiment of an optical image stabilization algorithm implemented in a driver integrated circuit of the system of Figure 7. Figure 9 illustrates more details of the driver integrated circuit of Figure 8; Figure 10 is a flow chart of the image stabilization procedure performed by the systems of Figures 8 and 9; Figure 11 is a system of Figure 7 A block diagram of one embodiment of an optical image stabilization algorithm implemented in an image processor integrated circuit; FIG. 12 illustrates more details of the driver and image processor integrated circuit of FIG. 11; and FIG. 13 is a diagram 11 and FIG. 12 is a flowchart of one of the image stabilization programs executed by the system.

1‧‧‧Actuator

2‧‧‧Actuator

3‧‧‧Actuator

100‧‧‧Image Stabilization Accessories/Image Stabilizer

105‧‧‧Center aperture

106‧‧‧Flexing pieces

110‧‧‧Circular mounting stage

112‧‧‧Fixed comb support

113‧‧‧Removable comb support

115‧‧‧ Direction

118‧‧‧Pore Center/Lens Center

120‧‧‧linear displacement

121‧‧‧Fixed part

122‧‧‧ movable part

125‧‧‧External framework

Claims (19)

A camera comprising: a plurality of integrally formed electrostatic actuators, each actuator configured to apply a force to at least one lens, wherein each actuator comprises a fixed portion, a movable portion and a latch a lock, the moveable portion being configured to move relative to the fixed portion between a first position at which the actuator is operable and a second position at which the actuator is inoperable, the latch being configured Latching the movable portion to the first position; an optical image stabilization (OIS) algorithm module configured to command the plurality of actuators to actuate the at least one in response to movement of the camera a lens; and a driver integrated circuit operative to drive the actuators in response to commands from the OIS algorithm module. The camera of claim 1, wherein each actuator is configured to apply a full force to the at least one lens, and wherein the OIS algorithm module is configured to command the plurality of actuators to respond to the camera The movement tangentially actuates the at least one lens. The camera of claim 2, further comprising: a plurality of position sensors corresponding to the plurality of actuators, each position sensor measuring an all-direction displacement of the corresponding actuator; and a translation The module is operable to translate the tangential displacements from the position sensors into one of the displacements of the lens, wherein the OIS algorithm module is also responsive to the lens displacement. The camera of claim 1, wherein the OIS algorithm module is integrated in the drive The actuator integrated circuit, and wherein the OIS algorithm module is configured to command the driver integrated circuit to apply an acceleration pulse to the movable portion to latch the movable portion to the first position. The camera of claim 1, further comprising: an imager configured to digitally acquire an image through the at least one lens; and an image processor integrated circuit operable to process the digitized image The OIS algorithm module is integrated in the image processor integrated circuit. The camera of claim 1, wherein the camera is integrated into a cellular phone. The camera of claim 6, wherein the actuators are electrostatic comb actuators. The camera of claim 7, further comprising: a circular stage that receives the at least one lens, wherein the plurality of actuators comprises three actuators symmetrically disposed outside the circular stage . The camera of claim 1, wherein the OIS algorithm module includes a Kalman filter for predicting one of the expected motions of the camera. The camera of claim 9, further comprising: a gyroscope, wherein the gyroscope measures one of the tilt and roll of the camera, and wherein the Kalman filter is based on the measured trim and roll Predicting one of the expected motions of the camera, wherein the OIS algorithm module is configured to be based on the Kalman filter The controller predicts a difference between the expected motion of the camera and an unintended jitter of the camera to command the plurality of actuators to actuate the at least one lens. The camera of claim 1, wherein the OIS algorithm module commands the plurality of actuators based on a relatively lower data rate, and wherein the camera further includes a command for a relative command based on the OIS algorithm modules The high data rate controls at least one controller of the actuators. A method for image stabilization, comprising: sensing a motion of a camera using a gyroscope; determining, based on the sensed motion, a lens actuation of one of the stabilized camera lenses; translating the desired lens actuation into Steeringly tangentially actuating; tangentially actuating the at least one lens responsively to the desired tangential actuation; and moving the plurality of tangential actuators prior to tangentially actuating the lens One of each movable portion is latched in a first position in which the tangential actuator is operable to actuate the lens; the sensed camera motion from the gyroscope is filtered to provide Determining one of the expected positions of the lens; sensing tangential actuation by the plurality of tangential actuators; translating the tangential actuations into a current lens position; and subtracting the position from the expected lens position The current lens position provides a lens position error due to unintended camera motion, wherein the desired lens actuation stabilizes the lens by addressing the unintended camera motion. The method of claim 12, wherein the filtering is a Kalman filter. The method of claim 12, wherein the filtering is a high pass filtering. The method of claim 12, wherein the tangential actuation is a MEMS electrostatic comb tangential actuation, and wherein the movable portion of each of the plurality of tangential actuators is applied by applying an acceleration pulse Moving to the first position to the movable portion to cause one of the movable portions to leverage a latch. An actuator device comprising: a stage elastically supported for movement in a plane, wherein the stage includes a planar mounting structure disposed on the plane, the planar mounting structure being adapted Supporting an optical element; three or more actuators, each of which is coupled to an outer periphery of the stage and operable to apply to the plane along the line when actuated and cut a linear force to one of the stages; a flexure between each of the actuators and the outer periphery of the stage; wherein at least two of the flexures are coupled to the outer periphery One of a common position; wherein when the linear forces are applied to the flexures, the linear forces are configured to cooperate to effect the stage in any translation and/or any direction of rotation in the plane Moving; and an outer frame that surrounds and supports the stage and the actuators. The actuator device of claim 16, wherein the stage, the actuators, and the outer frame are disposed substantially coplanar with one another. The actuator device of claim 16, wherein at least one of the flex members is a recurvated flexure, and wherein each flexure is coupled to one of the extended portions of the stage The extended portion extends along a line between the two actuators. The actuator device of claim 16, wherein at least one of the actuators comprises: a fixed frame; a moving frame elastically supported to reciprocate relative to the fixed frame; and a plurality of linear fingers The teeth, which are alternately attached to the fixed frame and the moving frame, are oriented to reciprocate toward each other and away from each other with respect to all directions of the stage.