Detection, correction and compensation of coupling effects of micro-electro-mechanical system shafts

Abstract

Embodiments of the present disclosure generally relate to detection, correction, and compensation of coupling effects of microelectromechanical system axes. An oscillator control system includes an oscillator structure configured to oscillate about a first rotational axis and a second rotational axis according to a Lissajous pattern, wherein oscillation about the second rotational axis applies a cross-coupling error to oscillation about the first rotational axis, and wherein the cross-coupling error varies according to a Lissajous position within the Lissajous pattern; and a driver circuit including a phase-locked loop (PLL) configured to condition a drive signal that drives oscillation about the first rotational axis. The PLL is configured to generate a PLL signal based on a phase error of oscillation about the first rotational axis. The PLL includes a compensation circuit configured to receive the PLL signal and the Lissajous position within the Lissajous pattern, apply a compensation value to the PLL signal to generate a compensated PLL signal for generating the drive signal based on the Lissajous position.

Description

Technical Field

[0001] The embodiments of this disclosure generally relate to the detection, correction, and compensation of coupling effects of axes in microelectromechanical systems (MEMS). Background Technology

[0002] The use of high frequencies to scan collimated beams across a two-dimensional field of view in a defined manner is relevant to a variety of applications, such as light detection and ranging (LiDAR) or projection purposes in augmented reality (AR) applications. Due to their compact size and compatibility with mass production, microelectromechanical scanners (e.g., MEMS scanning mirrors) are attracting significant attention as key components of the beam guiding unit in such systems.

[0003] To deflect light in two orthogonal directions, a beam guiding system can combine two separate MEMS devices on two separate chips, each serving one axis. The two MEMS chips are spatially arranged such that a beam deflected in the first direction by a first scanning mirror strikes a second scanning mirror, which provides deflection in the second (orthogonal) direction. Thus, two one-dimensional (1D) MEMS scanning mirrors (each with one degree of operational freedom about its respective scanning axis) can be used to guide light in two dimensions. Alternatively, a two-dimensional (2D) MEMS scanning mirror with two degrees of operational freedom about the two scanning axes in the same device can be used to guide light in two dimensions.

[0004] In both implementations, each of the two axes can be designed for either quasi-static (QS) or resonant (RES) operation. More specifically, all combinations of QS / QS, QS / RES, and RES / RES are possible. However, the frequencies at which a QS mirror can be operated (range 10–500 Hz, heavily dependent on size) are typically much lower than the resonant frequencies of the MEMS mirror (range 1 kHz–40 kHz, heavily dependent on size). Therefore, a QS / QS combination is practically meaningless because it cannot provide high-resolution scan patterns at an acceptable frame rate.

[0005] In both implementations, the movable mirror (rotor) is suspended from the semiconductor chip frame by springs that provide restoring force. To deflect light, the mirror needs to be rotatable about one (1D) or two (2D) operating axes. The MEMS mirror is rotated by an actuation mechanism. Common actuation schemes for MEMS mirrors are electrostatic actuation, piezoelectric actuation, and electromagnetic actuation. All of these can be used for both QS and RES actuation. Electrostatic actuation is typically achieved through an interdigitated out-of-plane comb-like drive structure that forms a capacitance that changes with the MEMS position.

[0006] Even assuming the rotor of a 1D mirror is perfectly rigid, the spring-suspended mirror body has six (rigid body) degrees of freedom (DOF): three translational and three rotational, each associated with an eigenmode oscillation. Ideally, in operation, a MEMS mirror performs deflection in only one of these DOFs (i.e., the DOF associated with the desired rotation). Therefore, the design goal of a MEMS mirror is to suppress all unwanted eigenmodes, meaning their eigenfrequency should be as high as possible. This is achieved by designing the springs in a way that allows them to operate only for the desired rotational motion while being as rigid as possible for all other translations and rotations.

[0007] There are two main possibilities regarding the definition of the two operating axes of a 2D MEMS scanner. Either there is a hierarchical definition of the axes, or the two axes are equivalent. The first case is achieved by using so-called gimbaled MEMS devices with universal mounting, while the second case is achieved by a tripod or quadruped design that results in (almost) degradation of the eigenfrequency due to symmetry.

[0008] The 2D beam guiding unit, initially implemented using two 1D scanners, has the disadvantage that system integration is not as compact as in the case of 2D MEMS because the two chips need to be arranged at an angle relative to each other. However, it has the advantage of defining motion better than a true 2D mirror.

[0009] To illustrate this, consider a 2D gimbaled MEMS scanner. A 2D MEMS scanner has an inner scanning axis (i.e., the endoscope rotor) and an outer scanning axis (i.e., the exoscope rotor), the outer scanning axis typically oscillates at a lower frequency than the inner scanning axis. Ignoring the imperfect stiffness of the (end)scope rotor and the gimbal frame, the scanner still has twelve rigid body degrees of freedom, with the endoscope body and gimbal frame each having three translational and three rotational degrees of freedom. Furthermore, the gimbal frame adds a significant amount of additional mass, which needs to be moved to deflect about the outer axis. Typically, this results in much poorer suppression of unwanted (parasitic) modes compared to a 1D mirror and leads to crosstalk behavior between the two axes.

[0010] Several mechanisms exist that could cause the two operating modes of a 2D gimbal MEMS scanner to affect each other.

[0011] For example, the eigenmodes of internal and external rotation are not pure rotations of the mirror body and gimbal frame, respectively. Instead, these eigenmodes, or normal modes, have a small mixture of rotations of the corresponding other objects about the same axis. For example, the operating eigenmode of the mirror body includes a small rotation of the gimbal frame about an axis of the (internal) mirror body that is perpendicular to the operating axis of the (external) gimbal frame. This motion is called the (parasitic) "pitch" motion of the gimbal frame, while the desired motion of the two axes is called the "roll" motion of the two objects.

[0012] As another example, the mirror body and gimbal frame can include reinforcing structures on the bottom side. These reinforcing structures shift the object's center of mass, causing a mismatch between the z-position of the rotation axis and the center of mass. When the object oscillates around the rotation axis, this rotational imbalance causes inertial forces (Euler forces, centrifugal forces) to act on the center of mass. This results in the excitation of translational degrees of freedom.

[0013] As another example, the endoscope body moves within a rotational reference frame of the gimbal frame. This results in a Coriolis force acting on the endoscope body, which can generate other rotational degrees of freedom, such as the pitch and yaw motions of the endoscope body.

[0014] To control the amplitude (i.e., the deflection or rotation angle around the axis) and phase, MEMS scanners typically cannot operate in an open loop. Instead, they must be actuated in a closed loop, for example, using a phase-locked loop (PLL). This requires a sensing scheme to acquire information about the current rotational state. This information typically needs to always know at least two quantities, such as amplitude and phase, or the current deflection angle and angular velocity. Particularly for two resonant scanners, which are operated to scan a constant Lissajous pattern at optimal grid resolution, control of the relative phase of the oscillations around the two axes is crucial.

[0015] The ability to accurately sense the positional information (e.g., rotation angle) of a scanning mirror about one or more of its scanning axes is crucial for accurate laser imaging. More accurate and precise positional information results in more accurate laser imaging, better scanning resolution, and better scanning pattern generation. For electrostatic MEMS mirrors, a self-sensing scheme is convenient because it requires no additional sensing devices besides actuators. For resonant MEMS mirrors, a current self-sensing scheme is convenient because the actuation of the MEMS mirror and the resulting oscillations cause periodic charging and discharging currents in the comb-shaped drive electrodes. These currents can be measured, and information about the scanner's state can be obtained by analyzing the recurrence times of different characteristics of these currents (e.g., zero-crossing or peak current). For example, suitable combinations of various comb-shaped drive arms can be formed, and the total current of such combinations can be analyzed. If such a combination is symmetrical (meaning equivalent symmetrical arms from the left and right sides of the rotation axis must be added), the zero-crossing of the total current occurs precisely at the same time as the mechanical zero-crossing of the associated rotating body.

[0016] The aforementioned mode coupling phenomenon and the associated parasitic motion of the scanner (i.e., the mirror body and / or gimbal frame) also alter the variable capacitance of the comb-driven capacitor, thus resulting in displacement currents in the same manner as the expected motion (two rotations of the mirror body and gimbal frame). These parasitic currents are added to the current from the expected motion, and the mechanical sources of these currents cannot be distinguished a priori. In the case of zero crossings, this means that the current zero crossing is out of sync with the mechanical zero crossing, either preceding or following it. In conventional methods, this would lead to incorrect determination of the state of the MEMS scanner about one or both axes, i.e., incorrect determination of the phase and / or amplitude of the MEMS scanner about the corresponding axes. For example, the result would be a poorly controlled scan pattern with degraded resolution, flicker, amplitude variations, etc., leading to instability in the spherical angle of the projected image.

[0017] Therefore, a component may be needed for detecting, correcting, and compensating for cross-coupling effects between the axes of a 2D MEMS mirror. For example, to improve the scanning operation of a 2D MEMS mirror, a component may be needed to compensate for mode coupling effects and motion mixtures originating from the sensing current superimposed on the sensing current of one 2D scanning axis onto another 2D scanning axis. Summary of the Invention

[0018] One or more embodiments provide an oscillator control system comprising an oscillator structure configured to oscillate simultaneously about a first rotation axis and a second rotation axis according to a Lissajous pattern, the Lissajous pattern being repeated frame-by-frame, wherein the oscillation about the second rotation axis applies a deterministic cross-coupling error to the oscillation about the first rotation axis, and wherein the deterministic cross-coupling error varies according to the Lissajous position of the oscillator structure within the Lissajous pattern; and a driver circuit configured to generate a first drive signal for driving the oscillator structure about the first rotation axis at a first drive frequency and to generate a second drive signal for driving the oscillator structure about the second rotation axis at a second drive frequency different from the first drive frequency. A second drive signal drives a second oscillator structure via a rotating axis to generate a Lissajous pattern. The driver circuit includes a phase-locked loop (PLL) configured to regulate the oscillations about a first rotating axis. The PLL is configured to generate a PLL signal based on a phase error of the oscillations about the first rotating axis. The PLL includes a compensation circuit configured to receive the PLL signal and a Lissajous position signal indicating a Lissajous position within the Lissajous pattern. The compensation circuit is configured to apply a compensation value to the PLL signal, thereby modifying the PLL signal to generate a compensated PLL signal for generating the first drive signal. The compensation circuit is configured to determine the compensation value based on the Lissajous position indicated by the Lissajous position signal.

[0019] One or more embodiments provide a method for controlling an oscillator, the method comprising driving an oscillator structure to oscillate simultaneously about a first rotation axis and a second rotation axis according to a Lissajous pattern repeated frame by frame, wherein the oscillation about the second rotation axis applies a deterministic cross-coupling error to the oscillation about the first rotation axis, wherein the deterministic cross-coupling error varies according to the Lissajous position within the Lissajous pattern, wherein driving the oscillator structure further comprises generating a first drive signal for driving the oscillator structure about the first rotation axis at a first drive frequency, and generating a second drive signal for driving a second oscillator structure about the second rotation axis at a second drive frequency different from the first drive frequency to generate a Lissajous pattern. The Lissajous pattern; adjusting the oscillation about a first rotation axis using a phase-locked loop (PLL), including generating a PLL signal based on the phase error of the oscillation about the first rotation axis; and compensating the PLL signal using a Lissajous position signal indicating the Lissajous position within the Lissajous pattern and compensation information including a first set of compensation values, wherein each compensation value in the first set of compensation values ​​is encoded to a different Lissajous position within the Lissajous pattern, wherein compensating the PLL signal further includes selecting a compensation value from the first set of compensation values ​​based on the Lissajous position indicated by the Lissajous position signal, and applying the selected compensation value from the first set of compensation values ​​to the PLL signal, thereby modifying the PLL signal to generate a compensated PLL signal for generating a first drive signal.

[0020] One or more embodiments provide an oscillator control system including an oscillator structure configured to oscillate simultaneously around a first rotation axis and a second rotation axis at different frequencies, wherein the oscillation around the second rotation axis applies a systematic phase error to the oscillation around the first rotation axis, and wherein the systematic phase error changes sinusoidally over time; and a driver circuit configured to generate a first drive signal for driving the oscillator structure around the first rotation axis and a second drive signal for driving a second oscillator structure around the second rotation axis, wherein the driver circuit includes a primary phase-locked loop (PLL) configured to regulate the oscillation around the first rotation axis, wherein the primary PLL includes a primary phase error detector configured to generate a phase error signal based on a phase deviation between the oscillation around the first rotation axis and a desired oscillation phase, and output the phase error signal, wherein the systematic phase error is superimposed on the phase error signal, wherein the primary PLL includes a compensation circuit configured to receive the phase error signal, copy the systematic phase error from the phase error signal, and subtract the copied systematic phase error from the phase error signal to generate a compensated phase error signal for generating the first drive signal.

[0021] One or more embodiments provide a method for controlling an oscillator structure, the method comprising driving the oscillator structure to oscillate simultaneously around a first rotation axis and a second rotation axis at different frequencies, wherein the oscillation around the second rotation axis applies a systematic phase error to the oscillation around the first rotation axis, and wherein the systematic phase error changes sinusoidally over time, wherein driving the oscillator structure further comprises generating a first drive signal for driving the oscillator structure around the first rotation axis and generating a second drive signal for driving a second oscillator structure around the second rotation axis; regulating the oscillation around the first rotation axis using a primary phase-locked loop (PLL), including generating a phase error signal based on a phase deviation between the oscillation around the first rotation axis and a desired oscillation phase and outputting the phase error signal, wherein the systematic phase error is superimposed on the phase error signal; and compensating the phase error signal, including copying the systematic phase error from the phase error signal and subtracting the copied systematic phase error from the phase error signal to generate a compensated phase error signal for generating the first drive signal. Attached Figure Description

[0022] The embodiments are described herein with reference to the accompanying drawings.

[0023] Figure 1 This is a schematic block diagram of a two-dimensional scanning system according to one or more embodiments;

[0024] Figure 2Signal diagrams of various signals generated by a MEMS driver based on mirror angle θ and / or position according to one or more embodiments are shown;

[0025] Figure 3 It is a schematic block diagram of a MEMS control system according to one or more embodiments;

[0026] Figure 4 The time-dependent deterministic zero-crossing sensing error within a Lissajous frame according to one or more embodiments is illustrated;

[0027] Figure 5A This is a schematic block diagram of an error processing block utilizing one or more lookup tables according to one or more embodiments;

[0028] Figure 5B This is a schematic block diagram of an error processing block utilizing one or more lookup tables according to one or more embodiments;

[0029] Figure 6A This is a schematic block diagram of an error processing block utilizing one or more lookup tables according to one or more embodiments;

[0030] Figure 6B This is a schematic block diagram of an error processing block utilizing one or more lookup tables according to one or more embodiments;

[0031] Figure 7 This is a schematic block diagram of an error processing block utilizing one or more lookup tables according to one or more embodiments;

[0032] Figures 8A-8D A schematic block diagram of a PLL for a MEMS control system according to one or more embodiments is shown; and

[0033] Figures 9A-9C A schematic block diagram of an error processing block of a MEMS control system according to one or more embodiments is shown. Detailed Implementation

[0034] In the following, various embodiments will be described in detail with reference to the accompanying drawings. It should be noted that these embodiments are for illustrative purposes only and should not be construed as limiting. For example, while an embodiment may be described as including multiple features or elements, this should not be construed as indicating that all such features or elements are necessary to implement the embodiment. Rather, in other embodiments, some features or elements may be omitted or replaced with alternative features or elements. Furthermore, in addition to the features or elements explicitly shown and described, other features or elements, such as conventional components of a sensor device, may be provided.

[0035] Features from different embodiments may be combined to form other embodiments, unless otherwise specifically indicated. Variations or modifications described with respect to one embodiment may also apply to other embodiments. In some cases, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring the embodiments.

[0036] Furthermore, equivalent or similar elements, or elements with equivalent or similar functions, are indicated by equivalent or similar reference numerals in the following description. Since identical or functionally equivalent elements are given the same reference numerals in the drawings, repeated descriptions of elements with the same reference numerals can be omitted. Therefore, the descriptions provided for elements with the same or similar reference numerals are interchangeable.

[0037] Unless otherwise stated, the connections or couplings between the elements shown in the accompanying drawings or described herein can be wire-based connections or wireless connections. Furthermore, such connections or couplings can be direct connections or couplings without additional intermediate elements, or indirect connections or couplings with one or more additional intermediate elements, provided that the general purpose of the connection or coupling (e.g., transmitting a signal or conveying information) remains essentially unchanged.

[0038] In this disclosure, expressions including ordinal numbers such as "first" and "second" can modify various elements. However, such elements are not limited to the expressions described above. For example, the expressions described above do not restrict the order and / or importance of elements. The expressions described above are only used to distinguish one element from other elements. For example, the first box and the second box represent different boxes, although they are both boxes. As another example, the first element can be called the second element, and similarly, the second element can be called the first element without departing from the scope of this disclosure.

[0039] The embodiments relate to optical transmitters and systems configured to transmit light beams or pulses according to a scanning pattern, and more specifically, according to a 2D scanning pattern such as a Lissajous scanning pattern. The light beams include visible light, infrared (IR) light, or other types of illumination signals. In some applications, the transmitted light can be backscattered back into the system by an object, where the backscattered light is detected by a sensor. The sensor can convert the received backscattered light into an electrical signal, such as a current signal or a voltage signal, which can be further processed by the system to generate object data and / or an image.

[0040] For example, in a LiDAR (Light Detection and Ranging) system, a light source transmits light pulses into the field of view, and the light is reflected from one or more objects via backscattering. Specifically, LiDAR is a direct time-of-flight (TOF) system where light pulses (e.g., a laser beam of infrared light) are emitted into the field of view, and an array of pixels detects and measures the reflected beams. For example, a photodetector array receives the reflections from an object illuminated by light. The time difference of return of each light pulse across multiple pixels in the pixel array can then be used to create a digital 3D representation of the environment or generate other sensor data.

[0041] Lissajous scanning (e.g., according to a Lissajous scanning pattern employing two scanning axes) can illuminate a scene in a continuous scanning manner. By emitting continuous light pulses in different scanning directions, an area called the field of view can be scanned, and objects within that area can be detected and imaged. Therefore, the field of view represents the scanning plane with a projection center. Lissajous scanning is also useful in other applications, such as electronic displays on which images are rendered (e.g., displays for augmented reality (AR) applications) and automotive headlights for guiding light.

[0042] Figure 1 This is a schematic block diagram of a 2D scanning system 100 according to one or more embodiments. The 2D scanning system 100 includes a single two-dimensional (2D) MEMS mirror 12xy for guiding or otherwise deflecting a light beam (pulse) according to a 2D scanning pattern (such as a Lissajous scanning pattern).

[0043] The MEMS mirror 12xy is a mechanically moving mirror (i.e., a MEMS micromirror) integrated on a semiconductor chip (not shown). The MEMS mirror according to the embodiments described herein is configured to oscillate via rotation about two resonant scanning axes (i.e., a 2D MEMS mirror) that are generally orthogonal to each other. The oscillation of the MEMS mirror along the scanning axes can be between two predetermined extreme deflection angles (e.g., + / - 15 degrees). For example, a Lissajous scanner is configured to control the guidance of the beam in two dimensions (e.g., the horizontal x-direction and the vertical y-direction).

[0044] The MEMS mirror 12xy includes a first resonant scanning axis 13x (e.g., an outer scanning axis) that enables the MEMS mirror 12xy to guide light in the x-direction and a second resonant scanning axis 13y (e.g., an inner scanning axis) that enables the MEMS mirror 12xy to guide light in the y-direction. The two axes of a single 2D MEMS mirror are controlled by different phase-locked loops (PLLs). In this way, a single MEMS mirror can guide a light beam received from the illumination unit 10 in both the x and y directions. As a result, the MEMS mirror 12xy can guide the light beam to the desired 2D coordinates (e.g., xy coordinates) in the field of view. The MEMS mirror 12xy can guide multiple light beams at different 2D coordinates of the Lissajous pattern.

[0045] The MEMS mirror 12xy is itself a nonlinear resonator (i.e., a resonant MEMS mirror) configured to oscillate at a resonant frequency around each of its "left" and "right" scanning axes, causing light reflected from the MEMS mirror to oscillate back and forth in the scanning direction of the corresponding scanning axis. Each resonant scanning axis has a nonlinear correlation with respect to the resonant frequency and the maximum deflection amplitude due to the rigidity of the mirror's suspension around that axis. As will be described in further detail below, different resonant frequencies can be used for each scanning axis 13x and 13y to define a Lissajous pattern.

[0046] The 2D scanning system 100 also includes an illumination unit 10 (i.e., a light transmitter) comprising at least one light source (e.g., at least one laser diode or light-emitting diode) configured to transmit a light beam (pulse) along a transmission path toward the MEMS mirror(s). The illumination unit 10 may sequentially transmit multiple light pulses according to a trigger signal received from the system controller 23.

[0047] The 2D scanning system 100 also includes a system controller 23 configured to control components of the scanning system. In some applications, such as LiDAR, the system controller 23 may also be configured to receive raw data from a light sensor (not shown) and perform processing on it (e.g., via digital signal processing) to generate object data (e.g., point cloud data). Therefore, the system controller 23 includes at least one processor and / or processor circuitry (e.g., comparator, TDC, ADC, and digital signal processor (DSP)) for processing the data, and a control circuitry (such as a microcontroller) configured to generate control signals.

[0048] System controller 23 is configured to generate a trigger signal for triggering illumination unit 10 to generate light pulses. Therefore, system controller 23 controls the timing of light pulse emission from illumination unit 10 via the trigger signal. System controller 23 is also configured to set the drive frequency of the MEMS mirror for each of its scanning axes and is capable of synchronizing the oscillations around the two scanning axes 13x and 13y.

[0049] The 2D scanning system 100 includes a MEMS driver 25x for driving a MEMS mirror 12xy around a first scanning axis 13x and a MEMS driver 25y for driving a MEMS mirror 12xy around a second scanning axis 13y. Each MEMS driver 25x, 25y actuates and senses the rotational position of the mirror about its respective scanning axis and provides position information of the mirror (e.g., tilt / deflection angle or degree of rotation about the rotation axis) to a system controller 23. Based on this position information, the laser source of the illumination unit 10 can be triggered by the system controller 23. Therefore, the higher accuracy of the MEMS mirror position sensing leads to more accurate and precise control of other components of the scanning system.

[0050] A driving voltage (i.e., an actuation or driving signal) is applied by a MEMS driver to the actuator structure of the MEMS mirror corresponding to its corresponding scanning axis to drive the MEMS mirror to oscillate around that scanning axis. The driving voltage may be referred to as high voltage (HV). The actuator structure may include interdigital electrodes made of interdigital mirror combs and frame combs, to which the driving voltage (i.e., an actuation or driving signal) is applied by the MEMS driver.

[0051] The driving voltage applied to the actuator structure generates a driving force, for example, between the interdigitated mirror comb and the frame comb, which in turn produces torque on the mirror body about its axis of rotation. The driving voltage can be switched on and off (HV on / off) or switched between high and low voltage levels, thereby generating an oscillating driving force. This oscillating driving force causes the mirror to oscillate back and forth between two extreme values ​​on its axis of rotation. The driving voltage can be a constant driving voltage, meaning that the driving voltage is the same when actuated (i.e., switched to on / high). However, it should be understood that the driving voltage is switched on and off to generate mirror oscillation. Depending on the configuration, this actuation can be adjusted or modified by adjusting the driving voltage off-time, the high voltage level of the driving voltage, the low voltage level of the driving signal, and / or the duty cycle.

[0052] MEMS drivers 25x and 25y can also measure and record the mirror frequency and current corresponding to their respective axes using the capacitance changes in the comb-driven rotor and stator of the actuator structure used to drive the MEMS mirror 12xy. As the MEMS mirror 12xy oscillates around its respective axis, the capacitance between the finger electrodes of that axis changes according to the mirror's rotational position around that axis. Each MEMS driver 25x and 25y is configured to measure the capacitance between its respective interdigitated electrodes, thereby determining the rotational position (i.e., angular position) of the MEMS mirror 12xy. By monitoring the associated capacitance, each MEMS driver 25x and 25y can detect the zero-crossing event and its timing of its respective axis, and can determine the tilt angle of the MEMS mirror 12xy around its receiving axis at any given time.

[0053] Each MEMS driver 25x and 25y can also use the measured capacitance to determine the mirror frequency of its respective axis and record this information in the memory at the MEMS driver or system controller 23. The MEMS drivers 25x and 25y may also include a processing circuit system, including at least one processor (e.g., an analog signal processing circuit system and / or a digital signal processing circuit system), which is configured to process the measurement information from the measurement circuit to detect the mirror zero crossing and compensate for any sensing errors (e.g., due to axis cross-coupling effects).

[0054] Alternatively or concurrently, the system controller 23 may receive and process measurement information from the measurement circuitry of the MEMS drivers 25x and 25y. Therefore, the system controller 23 may also include a processing circuitry system, including at least one processor (e.g., an analog signal processing circuitry system and / or a digital signal processing circuitry system), configured to process the measurement information from the measurement circuitry to assess the mechanical health and / or the state of the chip package of the MEMS mirror 12xy.

[0055] The sensing of the position of the MEMS mirror 12xy is performed based on detectors configured to measure capacitance. For example, as the MEMS mirror moves, the geometry of the finger structure changes, resulting in a change in the geometry of the capacitance. As the geometry of the capacitance changes, the capacitance itself also changes. Therefore, a particular capacitance directly corresponds to a specific position (i.e., tilt angle) of the MEMS mirror. By sensing the capacitance of the finger structure, the MEMS drivers 25x and 25y can monitor and track the oscillations of the mirror and determine a specific position of the MEMS mirror, including its zero crossings around its respective axes.

[0056] One way to measure capacitance is to measure the current flowing through the finger structure (e.g., total displacement current), convert the measured current into a voltage, and then further correlate the voltage with the capacitance and / or rotation angle. However, any method for measuring capacitance can be used. The direction of rotation (e.g., positive or negative, left to right or right to left, clockwise or counterclockwise, etc.) can also be detected by measuring the change in capacitance over time, where a positive or negative change indicates the opposite direction of rotation. The MEMS drivers 25x and 25y can also record the current and voltage measured during capacitance measurement. Therefore, increasing the accuracy of the reflector's position sensing can improve the overall accuracy of the scanning system.

[0057] Because the mirror is driven around the scanning axis at an oscillating frequency, when the mirror rotates in the first rotational direction (e.g., from left to right or clockwise), it crosses the zero position (i.e., 0°) at some point in time. Similarly, when the mirror rotates in the second rotational direction (e.g., from right to left or counterclockwise), it will cross the zero position at some point in time. These moments of crossing the zero position can be called zero-crossing events that occur at the zero-crossing time.

[0058] By sensing the rotational position of the MEMS mirror 12xy about its corresponding scanning axes 13x and 13y, the MEMS actuators 25x and 12y can sense zero-crossing (ZC) events of the MEMS mirror 12xy. A zero-crossing event is the moment when the MEMS mirror 12xy has a 0° rotation angle on its scanning axis. Specifically, it is the moment when the MEMS mirror 12xy is parallel to the frame or in a neutral position. The neutral position can also be referred to as the stationary position (e.g., when the MEMS mirror 12xy stops after the driving force is turned off). Since the MEMS mirror 12xy oscillates back and forth between two rotational directions (e.g., clockwise and counterclockwise), two zero-crossing events occur during scanning—one when the mirror oscillates in the first rotational direction and one when the mirror oscillates in the second rotational direction. It will also be understood that an angle-crossing event of another predefined angle can also be used instead of a zero-crossing event.

[0059] In some embodiments, the event time may correspond to a non-zero-crossing event. For example, the sensed rotation angle may be some other angle other than 0°. However, for illustrative purposes, the examples in this document will be described in the context of sensing a zero-crossing event.

[0060] MEMS drivers 25x and 25y are configured to detect each zero-crossing event and record the time of each event. This timing information (i.e., the measured zero-crossing time) can then be transmitted as position information to the system controller 23. Specifically, MEMS drivers 25x and 25y trigger a change in the output of the corresponding position signal (Position_L) at each zero-crossing event or angle crossing event.

[0061] Figure 2 A signal diagram of various signals generated by the MEMS driver 25 based on the mirror angle θ and / or position is shown, including a position signal (Position_L). For example, the position signal (Position_L) can be a pulse signal during which a first pulse transition (e.g., a falling edge transition) is triggered at a zero crossing when the mirror oscillates in a first rotation direction (e.g., from left to right), and a second pulse transition (e.g., a rising edge transition) is triggered at a zero crossing when the mirror oscillates in a second rotation direction (e.g., from right to left). Furthermore, the signal is "high" when the mirror is pointing in one direction (e.g., to the left) and "low" when the mirror is pointing in a second direction (e.g., to the right). Therefore, the position signal indicates zero-crossing events not only by triggering pulse transitions but also by indicating the directional tilt of the mirror. The frequency of the position signal decreases as the interval between zero-crossing events increases. Based on this position signal, both the phase and / or frequency of two or more position signals can be compared.

[0062] Alternatively, each MEMS driver 25x and 25y can generate a short pulse at each zero-crossing event, causing a pulse position signal (Position_L) to be output to the system controller 23. That is, the signal remains low (or high) between zero-crossing pulses. In this case, absolute phase information indicating the direction of mirror movement will not be available. Based on this position signal, the phase and / or frequency of two or more position signals can be compared.

[0063] Each MEMS driver 25x and 25y can send position information to the system controller 23, enabling the system controller 23 to use the position information to control the triggering of laser pulses in the illumination unit 10. The position information can also be used by the system controller as feedback information, allowing the system controller 23 to maintain stable operation of the MEMS mirror 12xy and synchronization with other MEMS mirrors via control signals provided to the MEMS drivers 25x and 25y.

[0064] Therefore, the scanning technique involves transmitting a light beam from a transmission mirror into the field of view, the mirror using two resonant scanning axes according to a Lissajous scanning pattern. The transmission mirror oscillates continuously in a resonant manner around each scanning axis, causing the light beam to be projected into the field of view, which moves across the field of view as the transmission mirror changes its transmission direction. Furthermore, additional conditions are set by the system controller 23 to generate the Lissajous scanning pattern. The following conditions are used to synchronize the drive around the two scanning axes according to the Lissajous pattern.

[0065] For the Lissajous pattern to reproduce itself periodically at a frame rate FR frequency [Hz], additional conditions regarding frequencies f1 and f2 must be satisfied, where f1 is the time-domain driving frequency of the MEMS mirror 12xy around the scan axis 13x, and f2 is the time-domain driving frequency of the MEMS mirror 12xy around the scan axis 13y. However, the oscillations around the two scan axes may be asynchronous and must be synchronized by the system controller 23. A new frame begins each time the Lissajous pattern restarts, which occurs when the phase difference between the mirror phase around the scan axis 13x and the mirror phase around the scan axis 13y is zero.

[0066] For example, the coordinates X and Y of the transmitted beam are parametrically defined as oscillatory behavior variables in the time domain according to the following equation:

[0067] X = sin(2π*f1*t) (1),

[0068] Y = sin(2π*f2*t) (2).

[0069] X is the x-coordinate corresponding to the rotation angle X of the MEMS mirror around the scanning axis 13x, and Y is the y-coordinate corresponding to the rotation angle Y of the MEMS mirror around the scanning axis 13y. The X and Y coordinates are sinusoidal functions dependent on the driving frequencies f1, f2 and time (t). The MEMS mirror 12xy is driven around two resonant scanning axes at constant scanning frequencies f1 and f2, respectively, with a defined frequency ratio (M / N) between the scanning frequencies f1 and f2, where M and N are different integers, making the ratio non-uniform.

[0070] To create a repeatable Lissajous pattern (frame) with a frame rate FR, system controller 23 is configured to apply synchronization and frequency tuning to MEMS drivers 25x and 25y via control signals. Since the frame rate FR is predefined, system controller 23 can use the predefined frame rate FR as the greatest common divisor for selecting frequencies f1 and f2. In other words, system controller 23 selects frequencies f1 and f2 such that the frame rate FR is their greatest common divisor.

[0071] The embodiments relate to detecting, correcting, and compensating for cross-coupling effects between the axes of a 2D MEMS mirror. For example, a component can be implemented to compensate for mode coupling effects and motion mixtures originating from one 2D scan axis that are superimposed on the sensing current of the other 2D scan axis. Additionally or alternatively, the drive signal (i.e., HV on / off) corresponding to each scan axis can be compensated to account for mode coupling effects and motion mixtures.

[0072] Figure 3This is a schematic block diagram of a MEMS control system 300 according to one or more embodiments. The MEMS control system 300 includes a driver circuitry combining a circuitry of MEMS drivers 25x and 25y (e.g., two PLLs) with a compensation circuitry for driving a MEMS mirror 12xy around its respective axes 13x and 13y. The compensation circuitry is provided to compensate for mode coupling effects and motion mixtures originating from one 2D scan axis that are superimposed on the other 2D scan axis. Specifically, the compensation circuitry is configured to compensate for deterministic (e.g., modeled) parasitic cross-coupling effects (errors), where the cross-coupling error is known based on the Lissajous scan position within a Lissajous frame. That is, each Lissajous scan position within a Lissajous frame has a corresponding known cross-coupling error stored in a system model generated during system testing.

[0073] The MEMS control system 300 includes a reference oscillator 30 that generates an oscillator signal (e.g., a system clock signal) at a predetermined frequency, a first frequency divider 31 that divides the predetermined frequency of the oscillator signal by an integer value X, a second frequency divider 32 that further divides the frequency output from the first frequency divider 31 by an integer value M to generate a first reference frequency f1 corresponding to the X scan axis 13x, and a third frequency divider 33 that further divides the frequency output from the first frequency divider 31 by an integer value N to generate a second reference frequency f2 corresponding to the Y scan axis 13y.

[0074] Therefore, components 30-33 generate two reference signals, Ref1 and Ref2. Reference signal Ref1 has a fixed reference frequency f1 and is used as a reference signal for the PLL that drives the MEMS driver 25x of the MEMS mirror 12xy around the scan axis 13x at the reference frequency f1. Reference signal Ref2 has a fixed reference frequency f2 and is used as a reference signal for the PLL that drives the MEMS driver 25y of the MEMS mirror 12xy around the scan axis 13y at the reference frequency f2. Reference frequencies f1 and f2 are frequency-controlled and have a predefined frequency ratio (M / N) that can generate a repeatable Lissajous scan pattern.

[0075] Furthermore, the reference signal Ref1 can be a pulse signal, where each signal pulse corresponds to the expected zero-crossing time of the MEMS mirror 12xy with respect to the scanning axis 13x. Similarly, the reference signal Ref2 can be a pulse signal, where each signal pulse corresponds to the expected zero-crossing time of the MEMS mirror 12xy with respect to the scanning axis 13y.

[0076] To compensate for deterministic parasitic cross-coupling effects, and more specifically, to track the Lissajous scan position of the MEMS mirror 12xy within a Lissajous frame, a phase detector 34, a Lissajous frame start detector 35, a first counter 36, and a second counter 37 are used. The phase detector 34 is configured to receive two reference signals Ref1 and Ref2, detect their respective phases, and provide this phase information to the Lissajous frame start detector 35. The Lissajous frame start detector 35 is configured to detect when the phase difference between the two phases is zero—indicating the start of a new Lissajous frame. In response to detecting a zero phase difference, the Lissajous frame start detector 35 is configured to generate a frame start signal (e.g., a signal pulse), which also serves as a reset signal for counters 36 and 37 to reset the counters to 1 (or to zero, where the counters count up to M-1 and N-1, respectively).

[0077] The first counter 36 is configured to generate a first counter value incremented to M, and the second counter 37 is configured to generate a second counter value incremented to N. Because the reference frequencies f1 and f2 have a predefined frequency ratio (M / N), when the first counter value equals M and the second counter value equals N, they will return to being in phase (i.e., their phase difference is zero). Therefore, at this time, the Lissajous frame start detector 35 should generate a signal pulse that resets counters 36 and 37 to 1 (or resets them to 0, where the counters count up to M-1 and N-1 respectively).

[0078] As a result, the first counter value represents the Lissajous position of the MEMS mirror 12xy within the Lissajous frame relative to the scan axis 13x, which is divided into M discrete scan positions (i.e., 1 to M Lissajous scan positions). Similarly, the second counter value represents the Lissajous position of the MEMS mirror 12xy within the Lissajous frame relative to the scan axis 13y, which is divided into N discrete scan positions (i.e., 1 to N Lissajous scan positions). The corresponding scan position is used to determine the compensation value to be applied by the compensation circuit to compensate for deterministic parasitic cross-coupling effects (errors). The compensation value applied in the PLL of scan axis 13x depends on the first counter value representing the Lissajous scan position on the X-axis. Similarly, the compensation value applied in the PLL of scan axis 13y depends on the second counter value representing the Lissajous scan position on the Y-axis.

[0079] The PLL scanning axis 13x includes a signal processing circuit 50 (i.e., a mirror sensor) with an electrical ZC readout circuit 51, a phase frequency detector 52, an error signal generator 53, an error processing block 54, a loop filter 55, and an error processing block 56. Error processing blocks 54 and 56 include a processing circuit system that receives multiple inputs and applies compensation values ​​to the PPL signal to generate a compensated PLL signal.

[0080] The PLL scanning axis 13y includes a signal processing circuit 60 with an electrical ZC readout circuit 61, a phase frequency detector 62, an error signal generator 63, an error processing block 64, a loop filter 65, and an error processing block 66. Error processing blocks 64 and 66 include a processing circuit system that receives multiple inputs and applies compensation values ​​to the PPL signal to generate a compensated PLL signal.

[0081] Signal processing circuit 50 is configured to measure the displacement current received from the actuator comb of the scanning axis 13x and detect zero-crossing events of the MEMS mirror 12xy relative to the scanning axis 13x. Electrical readout circuit 51 is configured to generate... Figure 2 The position signal (Position_L) shown is similar to the position signal (Position_L_X), which indicates the zero-crossing of the measurement relative to the scan axis 13x. Ideally, but impractically, the phase and frequency of the position signal (Position_L_X) will be the same as those of the reference signal Ref1.

[0082] Signal processing circuit 60 is configured to measure the displacement current received from the actuator comb of the scanning axis 13y and detect zero-crossing events of the MEMS mirror 12xy relative to the scanning axis 13y. Electrical readout circuit 61 is configured to generate... Figure 2 The position signal (Position_L) shown is similar to the position signal (Position_L_Y), which indicates the zero-crossing of the measurement relative to the scan axis 13y. Ideally, but impractically, the phase and frequency of the position signal (Position_L_Y) will be the same as those of the reference signal Ref2.

[0083] Phase frequency detector 52 is configured to receive a position signal (Position_L_X) and a reference signal Ref1 and detect their phase. Phase frequency detector 52 can perform its phase detection at the rising edge transition (i.e., 0 to 1 transition) of the position signal (Position_L_X), at the falling edge transition (i.e., 1 to 0 transition) of the position signal (Position_L_X), or at both the rising and falling edges.

[0084] Error signal generator 53 is an error detector that compares two phases and generates an error signal px(m) that depends on the phase difference between the two phases (i.e., the phase error of the position signal (Position_L_X)). More specifically, the phase error represents the difference between the measured zero-crossing time provided by the position signal (Position_L_X) and the expected zero-crossing time provided by the reference signal Ref1. For example, if the measured zero-crossing time and the expected zero-crossing time are equal, the phase error will be zero. Otherwise, the phase error signal will be a non-zero value, corresponding to the difference between the measured zero-crossing time and the expected zero-crossing time.

[0085] For each zero crossing of the x-axis of the MEMS mirror 12xy corresponding to the edge detection scheme of the phase frequency detector 52, an error signal px(m) is actively output. For example, if the phase frequency detector 52 detects phase during the rising and falling edge transitions, an error signal px(m) can be actively output for each pass of the MEMS mirror (i.e., from left to right or from right to left), and each zero crossing results in a value for the error signal px(m). However, due to the cross-coupling of the two scan axes, the sensing error is superimposed on the phase error. Therefore, the error signal px(m) includes the phase error plus the sensing error superimposed thereon. The sensing error is deterministic based on the Lissajous scan position of the MEMS mirror 12xy relative to the scan axis 13x within the Lissajous frame. In other words, the system 300 knows the sensing error based on the first counter value output by the first counter 36.

[0086] Phase frequency detector 62 is configured to receive a position signal (Position_L_Y) and a reference signal Ref2 and detect their phase. Phase frequency detector 62 can perform its phase detection at the rising edge transition (i.e., 0 to 1) of the position signal (Position_L_Y), at the falling edge transition (i.e., 1 to 0) of the position signal (Position_L_Y), or at both the rising and falling edges.

[0087] Error signal generator 63 is an error detector that compares two phases and generates an error signal py(n) that depends on the phase difference between the two phases (i.e., the phase error of the position signal (Position_L_Y)). More specifically, the phase error represents the difference between the measured zero-crossing time provided by the position signal (Position_L_Y) and the expected zero-crossing time provided by the reference signal Ref2. For example, if the measured zero-crossing time and the expected zero-crossing time are equal, the phase error will be zero. Otherwise, the phase error signal will be a non-zero value, corresponding to the difference between the measured zero-crossing time and the expected zero-crossing time.

[0088] For each zero-crossing of the MEMS mirror 12xy corresponding to the edge detection scheme of the phase frequency detector 62, an error signal py(n) is actively output. For example, if the phase frequency detector 62 detects phase during the rising and falling edge transitions, an error signal px(m) can be actively output for each pass of the MEMS mirror (i.e., from left to right or from right to left), and each zero-crossing results in a value for the error signal py(n). However, due to the cross-coupling of the two scan axes, the zero-crossing sensing error is superimposed on the phase error. Therefore, the error signal py(n) includes the phase error plus the zero-crossing sensing error superimposed on it. The zero-crossing sensing error is based on the deterministic Lissajous scan position of the MEMS mirror 12xy relative to the scan axis 13y within the Lissajous frame. In other words, the system 300 knows the zero-crossing sensing error based on the counter value output by the second counter 37.

[0089] It should be noted that the frame start signal, position signal (Position_L_X), and position signal (Position_L_Y) can be output from the MEMS control system 300 to the system controller 23, for example, to adjust the timing of the light pulses emitted at the MEMS mirror system.

[0090] Error processing blocks 54, 56, 64 and 66 are integrated into their respective PLLs to compensate for one or more error sources (i.e., deterministic parasitic cross-coupling effects (errors)).

[0091] One type of error source relates to the problem of one axis affecting the sensing of a second axis. In this case, compensation is needed for zero-crossing sensing errors in position provisioning. In other words, this type of error is superimposed on the displacement current, thus affecting the timing of the measured (detected) ZC event. This, in turn, adds an additional error (i.e., zero-crossing sensing error) to the phase errors of the error signals px(m) and py(n). To compensate, it is not necessary to change the actual mirror motion (i.e., it is not necessary to change the energy injection of the MEMS driver). Instead, the first error processing block of each PLL (i.e., error processing blocks 54 and 64) compensates its corresponding error signal to generate a compensated error signal that describes the zero-crossing sensing error.

[0092] Figure 4A time-dependent deterministic zero-crossing sensing error within a Lissajous frame according to one or more embodiments is illustrated. In other words, the deterministic zero-crossing sensing error depends on or is determined by the Lissajous position of the corresponding scan axis of the 2D MEMS mirror 12xy. In this case, the deterministic zero-crossing sensing error is the zero-crossing delay time relative to a measured zero-crossing (e.g., relative to an expected zero-crossing) introduced by the cross-coupling effect of another scan axis. The deterministic zero-crossing sensing error is plotted or modeled over the duration of the Lissajous frame and repeated on a frame-by-frame basis. Such a zero-crossing delay time waveform is measured during system testing and stored as a deterministic system model.

[0093] System testing can be performed during manufacturing or periodically throughout the system's lifecycle to update the deterministic system model. Model and system data can be acquired, at least partially, through a calibration process that individually operates each axis (thus, no cross-coupling is given), and then the parameters of interest for the actual error correction process are calculated or learned. Detailed analysis of the spacing mixture case shows that the superimposed zero-crossing sensing error is frequency-dependent on the axis and therefore related to the frame rate of the Lissajous pattern. By understanding this deterministic behavior, digital circuit systems can be designed that utilize deterministic error sources to compensate for measured ZC time events.

[0094] By tracking the Lissajous scan positions of the corresponding scan axes (e.g., via counter 36 or counter 37), a deterministic zero-crossing sensing error (e.g., zero-crossing delay time) can be determined for each Lissajous scan position, and a compensation value corresponding to the deterministic zero-crossing sensing error can be applied to the error signal at each zero-crossing. The compensation value compensates for the deterministic zero-crossing sensing error (e.g., zero-crossing delay time). This compensation is performed by error processing blocks 54 and 64 for each scan axis. Thus, for example, a deterministic system model is stored at system controller 23 for each scan axis 13x and 13y to compensate for the deterministic zero-crossing sensing error caused by the cross-coupling effect of another scan axis.

[0095] Error processing block 54 receives error signal px(m) (i.e., the phase error of the measurement of the zero-crossing event relative to the Lissajous reference signal Ref1), the current measurement of the zero-crossing event POSITION_L_X of the scan axis 13x, the Lissajous position within the Lissajous pattern (e.g., represented by the counter value from counter 36), and the model and system data of the entire Lissajous system provided by system controller 23, which may include MEMS model data (e.g., deterministic zero-crossing sensing error plotted relative to the X-axis over time or with the Lissajous position), including knowledge about Coriolis force, mixture spacing effect, etc.

[0096] The model system and data can be received by error processing block 54 and stored in one or more lookup tables (LUTs). Specifically, knowledge regarding the relationship between the Lissajous position of the MEMS mirror within a Lissajous frame and the spacing modulation error on the zero-crossing detection event of the scan axis is stored as part of the model and system data. Each lookup table encodes an error compensation value to each corresponding Lissajous position, indicated, for example, by counter 36. In other words, an error compensation value is assigned to each Lissajous position 1-M such that the error compensation value is a function of the Lissajous position (i.e., error_compensation(m) = f(Lissajous_position(m)), where m is the current Lissajous position of the scan axis 13x).

[0097] Based on this input data, error processing block 54 acts as a filter structure that modifies (i.e., compensates) the position supply. For example, the mixed-pitch effect is a deterministic effect that modulates the zero-crossing sensing of the outer axis based on the inner axis without altering the actual mirror motion. Therefore, error processing block 54 generates a compensated error signal px(m)' according to the following equation: paid_phase_error(m) = phase_error(m) - error_compensation(m), thereby eliminating the effects of pitch cross-coupling from the calculated error signal px(m) and thus from the position supply.

[0098] Figure 5A This is a schematic block diagram of an error processing block 54 utilizing one or more lookup tables according to one or more embodiments. Here, two lookup tables, LUT1x and LUT2x, are shown, which are compensators based on the cross-coupling effects of different axes at Lissajous positions 1 to M, indicated by counter 36. The two lookup tables, LUT1x and LUT2x, are filled with compensation values ​​provided by model and system data as a function of Lissajous positions 1 to M. Each LUT entry 1 to M is mapped to a different Lissajous position 1 to M indicated by counter 36.

[0099] Therefore, the first lookup table LUT1x is configured to receive Lissajous positions (i.e., counter values) from counter 36 and output compensation values ​​from the LUT entries corresponding to the received Lissajous positions. For example, if the counter value is 2 (i.e., Lissajous position 2 from 1 to M), the compensation value from LUT_entry_2 is output from the first lookup table LUT1x. The compensation value from the first lookup table LUT1x is provided to a signal compensator 54a (e.g., a subtractor or adder) that subtracts (or adds) the compensation value from the error signal px(m).

[0100] The position signal (Position_L_X) can be used to trigger the LUT lookup operation. For example, the error processing block 54 can be triggered by the same edge detection scheme used by the phase frequency detector 52. For example, if the phase frequency detector 52 is triggered by the rising edge detection of the position signal (Position_L_X) to output the error signal px(m), then the first lookup table LUT1x will similarly be triggered by the rising edge detection of the position signal (Position_L_X) to output the compensation value. In this way, the output of the lookup table can be synchronized with the generation of the error signal px(m).

[0101] Similarly, the second lookup table LUT2x is configured to receive Lissajous positions (i.e., counter values) from counter 36 and output compensation values ​​from the LUT entries corresponding to the received Lissajous positions. A position signal (Position_L_X) can be used to trigger the LUT lookup operation. The compensation value from the second lookup table LUT2x is provided to a signal compensator 54b (e.g., a subtractor or adder) that subtracts (or adds) the compensation value from the error signal px(m).

[0102] It should be noted that, given the deterministic and periodic nature of the cross-coupling effect, LUT entries follow the rule: Sum(LUT_Entry_1,2,…M) = 0. In other words, the sum of all compensation values ​​stored in LUT1x is equal to 0, and the sum of all compensation values ​​stored in LUT2x is equal to 0.

[0103] Error processing block 54 then outputs the compensated error signal px(m)' to loop filter 55. Loop filter 55 changes the properties of mirror driver 25x (e.g., voltage level, phase, duty cycle, etc.) according to the compensated error signal px(m)'.

[0104] Error processing block 64 receives error signal py(n) (i.e., the phase error of the measurement of the zero-crossing event relative to the Lissajous reference signal Ref2), the current measurement of the zero-crossing event POSITION_L_Y of the scan axis 13y, the Lissajous position within the Lissajous pattern (e.g., represented by the counter value from counter 37), and the model and system data of the entire Lissajous system provided by system controller 23, which may include MEMS model data (e.g., deterministic zero-crossing sensing error plotted relative to the Y-axis over time or with the Lissajous position), including knowledge about Coriolis force, mixture spacing effect, etc.

[0105] The model system and data can be received by error processing block 64 and stored in one or more lookup tables (LUTs). Specifically, knowledge regarding the relationship between the Lissajous position of the MEMS mirror within a Lissajous frame and the spacing modulation error on the zero-crossing detection event of the scan axis is stored as part of the model and system data. Each lookup table encodes an error compensation value to each corresponding Lissajous position, indicated, for example, by counter 37. In other words, an error compensation value is assigned to each Lissajous position 1-N such that the error compensation value is a function of the Lissajous position (i.e., error_compensation(n) = f(Lissajous_position(n)), where n is the current Lissajous position of the scan axis 13y).

[0106] Based on this input data, error processing block 64 acts as a filter structure that modifies (i.e., compensates) the position supply. For example, the hybrid spacing effect is a deterministic effect that modulates the zero-crossing sensing of the outer axis based on the inner axis without altering the actual mirror motion. Therefore, error processing block 64 generates a compensated error signal py(n)' according to the following equation: paid_phase_error(n) = phase_error(n) – error_compensation(n), thereby eliminating the effects of spacing cross-coupling from the calculated error signal py(n) and thus from the position supply.

[0107] Figure 5B This is a schematic block diagram of an error processing block 64 utilizing one or more lookup tables according to one or more embodiments. Here, two lookup tables, LUT1y and LUT2y, are shown, which are compensators for the cross-coupling effects of different axes based on the Lissajous positions 1-N indicated by counter 37. The two lookup tables LUT1y and LUT2y are filled with compensation values ​​provided by model and system data as a function of the Lissajous positions 1-N. Each LUT entry 1-N is mapped to a different Lissajous position 1-N indicated by counter 37.

[0108] Therefore, the first lookup table LUT1y is configured to receive Lissajous positions (i.e., counter values) from counter 37 and output compensation values ​​from the LUT entries corresponding to the received Lissajous positions. For example, if the counter value is 2 (i.e., Lissajous position 2 from 1 to N), the compensation value from LUT_entry_2 is output from the first lookup table LUT1y. The compensation value from the first lookup table LUT1y is provided to a signal compensator 64a (e.g., a subtractor or adder) that subtracts (or adds) the compensation value from the error signal py(n).

[0109] The position signal (Position_L_Y) can be used to trigger the LUT lookup operation. For example, the error processing block 64 can be triggered by the same edge detection scheme used by the phase frequency detector 62. For example, if the phase frequency detector 62 is triggered by the rising edge detection of the position signal (Position_L_Y) to output the error signal py(n), then the first lookup table LUT1y will similarly be triggered by the rising edge detection of the position signal (Position_L_Y) to output the compensation value. In this way, the output of the lookup table can be synchronized with the generation of the error signal py(n).

[0110] Similarly, the second lookup table LUT2y is configured to receive Lissajous positions (i.e., counter values) from counter 37 and output compensation values ​​from the LUT entries corresponding to the received Lissajous positions. A position signal (Position_L_Y) can be used to trigger the LUT lookup operation. The compensation value from the second lookup table LUT2y is provided to a signal compensator 64b (e.g., a subtractor or adder) that subtracts (or adds) the compensation value from the error signal py(n).

[0111] It should be noted that, given the deterministic and periodic nature of the cross-coupling effect, LUT entries follow the rule: Sum(LUT_Entry_1,2,…N) = 0. In other words, the sum of all compensation values ​​stored in LUT1y is equal to 0, and the sum of all compensation values ​​stored in LUT2y is equal to 0.

[0112] Error processing block 64 then outputs the compensated error signal py(n)' to loop filter 65. Loop filter 65 changes the properties of mirror driver 25y (e.g., voltage level, phase, duty cycle, etc.) according to the compensated error signal py(n)'.

[0113] Another type of error source relates to the problem of one axis affecting the motion of another axis (e.g., due to the Coriolis effect). In this case, compensation for the actuation of the other axis is required (i.e., a change in energy injection is needed). These problems are compensated by corresponding error processing blocks 56 and 66 in a manner similar to that described above with reference error processing blocks 54 and 64, except that error processing blocks 56 and 66 receive the outputs of loop filters 55 and 65, respectively, and perform compensation on the loop filter outputs LFx and LFy. Error processing blocks 56 and 66 change the properties of their mirror drivers 25x and 25y (e.g., voltage level, phase, duty cycle, etc.) according to their compensation. Naturally, the system model and data received by error processing blocks 56 and 66 are also different from the system model and data received by error processing blocks 54 and 64 for addressing different error sources and corresponding to the loop filter outputs opposite to the phase error signal. For example, compensation values ​​can be used to compensate for the voltage level, phase, or duty cycle of the drive signal.

[0114] Figure 6A This is a schematic block diagram of an error processing block 56 utilizing one or more lookup tables according to one or more embodiments. Error processing block 56 receives the loop filter output LFx from loop filter 55 and performs compensation on it to generate a compensated loop filter output LFx'. The lookup and compensation operations are similar to those described above. Figure 5A The operation described differs from the above-mentioned differences. One or more signal compensators 56a and 56b are used to add or subtract a compensation value to the loop filter output LFx to generate a compensated loop filter output LFx'. The compensated loop filter output LFx' is provided to the mirror driver 25x to adjust its drive signal, while cross-coupling effects are compensated (e.g., eliminated).

[0115] Figure 6B This is a schematic block diagram of an error processing block 66 utilizing one or more lookup tables according to one or more embodiments. The error processing block 66 receives the loop filter output LFy from the loop filter 65 and performs compensation on it to generate a compensated loop filter output LFy'. The lookup and compensation operations are similar to those described above. Figure 5A The operation described differs from the above-mentioned differences. One or more signal compensators 66a and 66b are used to add or subtract a compensation value to the loop filter output LFy to generate a compensated loop filter output LFy'. The compensated loop filter output LFy' is provided to the mirror driver 25y to adjust its drive signal, while cross-coupling effects are compensated (e.g., eliminated).

[0116] Another type of error source relates to the problem of one axis affecting another, but this effect cannot be sensed or measured like the spacing mixture effect and the Coriolis effect. In other words, the error is known to exist but cannot be sensed. However, this effect can be added to / subtracted from the error value, thus allowing the error value to be corrected. Furthermore, it may be necessary to compensate for the actuation of the other axis (change in energy injection). These types of errors can be compensated in error processing blocks 54 and 64, error processing blocks 56 and 66, or all error processing blocks.

[0117] Figure 7This is a schematic block diagram of an error processing block 54 utilizing one or more lookup tables according to one or more embodiments. Here, one or more compensation lookup tables 71 are used to compensate for deterministic and periodic effects. Furthermore, one or more correction lookup tables 72 with stored correction parameters are used to add errors to the measured error signal px(m) according to the Lissajous positions 1 to M. As previously described, each lookup table is populated based on the model system and data such that each lookup table encodes an error compensation value or error correction value to each corresponding Lissajous position and subsequently outputs a value based on the Lissajous position indicated by counter 36. Signal compensators 54a and 5b are used to add or subtract the compensation value or correction value to the error signal px(m) to generate compensated and corrected error signals px(m)'.

[0118] Another embodiment introduces an integration method with respect to several Lissajous frames, as shown in the following equation. In this embodiment, the LUT entry (m) from the current Lissajous frame is updated with the LUT entry (m-1) from the previous Lissajous frame and a reduced compensation value after low-pass filtering (LPF).

[0119] LUT Entry_m=LUT Entry_(m-1)-Compensation_(m-1) / 2^LPF

[0120] According to another embodiment, f1 / f2 = m / n and m and n can be any numbers (e.g., non-integers) and the frequencies f1 and / or f2 do not need to be controlled, so they do not need to remain constant over time.

[0121] Figures 8A-8D A schematic block diagram of a PLL of a MEMS control system 300 according to one or more embodiments is shown. Although Figures 8A-8D This relates to a PLL for controlling the MEMS mirror 12xy around the scanning axis 13x, but a similar PLL can be applied to a PLL for controlling the MEMS mirror 12xy around the scanning axis 13y. Therefore, it will be understood that... Figures 8A-8D The PLL shown can be similarly applied to Figure 3 The two PLLs shown drive the MEMS mirror 12xy around two axes. It should also be noted that the error generator 53 is integrated into the phase frequency detector (PFD) 52, and the PFD 52 outputs an error signal px(m). The measured frequency of the mirror (i.e., the mirror frequency and the frequency of the position signal Position_L) is denoted as fmir.

[0122] In this example, it can be assumed that the reference frequency f1 / f2 = M / N, and M and N can be any numbers (e.g., non-integers). Furthermore, there is no need to control the reference frequencies f1 and / or f2. Therefore, they do not need to remain constant over time.

[0123] The PLL includes: a mirror sensor 50 that detects the zero-crossing event as described above and generates a position signal Position_L with a measured mirror oscillation frequency fmir; a PFD 52 that receives both the position signal Position_L and the reference signal Ref1 and generates an error signal px(m) (or py(n)); a loop filter 55 that generates a loop filter signal LFx; a mirror driver 25x configured as a mirror scheduler that determines the sub-time (e.g., phase and duty cycle) period of the MEMS mirror 12xy around the scanning axis 13 and determines the low and high voltage levels of the drive signal based on the loop filter signal LFx; and a MEMS mirror 12xy that completes the loop.

[0124] The mirror driver 25x includes at least one processor for performing time analysis and scheduling. Based on sub-time, the mirror driver 25x is configured to generate a drive voltage HV as a control signal for controlling the oscillation of the MEMS mirror 12xy around the scan axis 13x. More specifically, the drive voltage switches between at least two voltage levels (e.g., on and off, or high and low voltage levels) and is applied to the actuator structure of the MEMS mirror 12xy described above.

[0125] like Figure 8A As shown, PFD 52 can respond to one or both transition edges of the position signal Position_L. In undisturbed (ideal) operation, the measured mirror oscillation frequency fmir has the same frequency and constant (ideally 0) phase as the reference frequency f1, resulting in no change to the PFD output (i.e., the error signal px(m)) of the loop filter output LFx, allowing the mirror driver 25x to drive the MEMS mirror 12xy with a drive signal having a fixed pulse sequence. Note that capacitively actuated mirrors are typically driven at twice the mirror oscillation frequency fmir.

[0126] like Figure 8B As shown, if the mirror oscillation is affected by any interference, the measured mirror oscillation frequency fmir will deviate from the reference frequency f1, causing the mirror phase error signal px(m) to deviate from zero, which will be regulated by the PLL. The PFD 52 detects the deviation, the loop filter 55 filters the error signal px(m), and changes the properties (voltage, phase, duty cycle, etc.) of the mirror driver output signal according to the detected phase error. Assuming the PLL is correctly parameterized, the mirror oscillation is regulated by the PLL, thereby eliminating interference, and fmir is synchronized with the reference frequency f1 again.

[0127] like Figure 8CAs shown, if the mirror oscillation is systematically affected by, for example, a second axis oscillating at a different frequency, the phase of the measured mirror oscillation frequency fmir will systematically deviate from the phase of the reference frequency f1 in a sinusoidal manner or by an overlap of several sinusoidal curves, for example, as Figure 4 As shown. The PFD 52 detects phase deviations, and the PLL continuously attempts to adapt the mirror oscillations to periodic disturbances. These periodic (sinusoidal) disturbances are called deterministic errors, and the disturbances caused by other axes of the 2D MEMS mirror are called deterministic cross-coupling effects or errors, which can be evaluated in system testing and stored as deterministic models and system data based on Lissajous positions. These periodic disturbances lead to unnecessary adjustment work, and the portion of the systematic phase error and oscillation inertia not filtered by the loop filter 55 will be visible in the mirror oscillations.

[0128] like Figure 8D As shown, an error processing block 54 is added to the PLL between the PFD 52 and the loop filter 55. The error processing block 54 compensates for systematic phase error signals (e.g., sensing errors) present in the error signal px(m) to generate a compensated error signal px(m)'. As described above, the error processing block 54 can be implemented as one or more lookup tables, or as a PLL.

[0129] Figures 9A-9C A schematic block diagram of an error processing block 54 of a MEMS control system 300 according to one or more embodiments is shown. Although Figures 9A-9C Error processing blocks 54 are involved, but they can be similarly applied to any error processing block disclosed herein. Furthermore, while error processing blocks 54 can be implemented using one or more lookup tables, for example, such as... Figure 5A As shown, but Figures 9A-9C The error processing block 54 is implemented using one or more dedicated PLLs. Each dedicated PLL follows the parasitic effects over a period of time. Therefore, each dedicated PLL is configured to compensate for different periodic (sinusoidal) parasitic effects that cause deterministic systematic errors, or can compensate for different spectral components of deterministic systematic errors. Important values ​​(such as phase and amplitude) can be determined via a learning phase and refined and adjusted during operation (i.e., throughout the entire lifetime of the MEMS mirror 12xy).

[0130] like Figure 9AAs shown, error processing block 54 is configured with appropriate compensation parameters, such as frequency, amplitude, and phase relative to the position within a Lissajous frame. In the case of the entire circuit system 901 to 906 being implemented digitally, error processing block 54 also receives a clock signal. If integrated on a chip, the digital implementation occupies less area than the analog implementation. Error processing block 54 can also output current compensation values ​​to system controller 23, which can monitor the compensation values ​​and adjust the configuration parameters based on the monitoring. The configuration parameters can be used to configure one or more phase loop filters and one or more amplitude loop filters.

[0131] like Figure 9B As shown, the error processing block 54 includes two control loops—a compensation PLL for phase and frequency control of the voltage-controlled oscillator (VCO) and an amplitude control loop for amplitude control of the voltage-controlled oscillator (VCO).

[0132] The compensation PLL includes PFD 901, phase loop filter 902, and VCO 903. The phase error between the phase of the PFD 901 output error signal px(m) and the phase of the VCO 903 output signal VCOout is calculated.

[0133] The amplitude control loop includes an amplitude detector 904, an amplitude loop filter 905, and a VCO 903. The amplitude error between the amplitude of the error signal px(m) output by the amplitude detector 904 and the amplitude of the output signal VCOout of the VCO 903 is measured.

[0134] The compensation PLL adjusts the phase of VCO 903 to synchronize its phase with the systematic periodic (sinusoidal) phase error present in the error signal px(m). Once phase synchronized, the amplitude control loop adjusts the amplitude of VCO 903 to the correct value, causing the VCO output signal VCOout to replicate the systematic phase error at the compensated error signal px(m). In other words, VCO 903 generates an output signal VCOout that replicates the periodic error signal included in the error signal px(m), which is caused by a periodic (sinusoidal) parasitic effect. The amplitude and phase of the VCO output signal VCOout are matched with the amplitude and phase of the periodic error signal through the control loop.

[0135] A periodic error signal is superimposed on the mirror phase error detected by PFD 52. By replicating the periodic error signal in both phase and amplitude, the periodic error signal (i.e., VCOout) can be subtracted from the error signal px(m) by the signal compensator 906 (e.g., a subtractor) to generate a compensated error signal px(m)'. Subtracting the VCO output signal VCOout from the error signal px(m) produces a constant signal instead of a periodic signal at the output of the signal compensator 906. It is also envisioned that other methods for matching phase and amplitude could be used to replicate systematic phase errors.

[0136] like Figure 9C As shown, the error processing block 54 includes two pairs of parallel-connected and operating control loops. Each pair of control loops has a compensation PLL for phase and frequency control of its respective voltage-controlled oscillator (VCO) and an amplitude control loop for amplitude control of its respective VCO. Each control loop is configured to compensate for different spectral components of the error signal px(m). As described above, the error signal px(m) may include a superposition of several sine curves (i.e., several periodic spectral error components).

[0137] The first control loop includes Figure 9B The control loop described herein, wherein the phase loop filter 902 and the amplitude loop filter 905 are parameterized such that the loop bandwidth is small enough to respond only to the target spectral component of the systematic phase error to compensate for spectral component #1. The VCO output signal of VCO 903 is a copy of the target spectral component of the systematic phase error.

[0138] The second control loop pair is similar to the first control loop pair, but is parameterized such that the loop bandwidth is small enough to respond only to different target spectral components of the systematic phase error to compensate for spectral component #2. The second control loop pair includes a compensation PLL and an amplitude control loop. The compensation PLL includes a PFD 907, a phase loop filter 908, and a VCO 909. The amplitude control loop includes an amplitude detector 910, an amplitude loop filter 911, and a VCO 909. The compensation PLL adjusts the phase of the VCO 909 to synchronize the phase of the VCO 909 with the phase of the target systematic periodic (sinusoidal) phase error present in the error signal px(m). Once phase synchronized, the amplitude control loop adjusts the amplitude of the VCO 909 to the correct amplitude value, such that the VCO output signal of the VCO 909 replicates the second target spectral component of the systematic phase error at the compensated error signal px(m)'. In other words, VCO 909 generates an output signal that replicates another target spectral component of the periodic error signal within a predetermined bandwidth of the error signal px(m), where the target spectral component is caused by the target periodic (sinusoidal) parasitic effect. The amplitude and phase of the VCO output signal are matched with the amplitude and phase of the second target spectral component of the systematic phase error through a control loop.

[0139] As a result of two PLL pairs, the two distinct spectral components of the periodic error signal are copied and subsequently subtracted from the error signal px(m) by the signal compensator 906 to generate the compensated error signal px(m)', i.e., there are no two distinct spectral components.

[0140] The detector, loop filter, VCO, and subtractor can preferably be implemented digitally. The digital implementation of the VCO (i.e., digitally controlled oscillator) can be implemented as a counter. The loop filters 902, 905, 908, and 911 can be parameterized via configuration parameters such that their loop bandwidth is small enough to react only to the target systematic phase error or its spectral components, and not to random phase errors caused by disturbances. If the systematic phase error includes more than one significant spectral component, a suitable number of error processing PLL pairs can be used, each pair compensating for a specific spectral component of its systematic phase error.

[0141] It will also be understood that the properties of the MEMS mirror 12xy can change with temperature or during the mirror's lifetime, for example, due to aging effects. Therefore, it is natural that the systematic phase error can also change with temperature or during the mirror's lifetime. The system controller 23 can receive feedback information (e.g., compensation values) from the error processing block 54 and can continuously or periodically adjust the parameters of the loop filters 902, 905, 908, and 911 to account for temperature variations and aging effects. Furthermore, error processing blocks exceeding defined frequency or amplitude windows can indicate unsafe degradation of the MEMS mirror 12xy and can forward this finding to functional safety mechanisms, such as the system controller 23.

[0142] Other embodiments are provided below.

[0143] 1. An oscillator control system comprising an oscillator structure configured to oscillate about a first rotation axis and a second rotation axis according to a Lissajous pattern, wherein oscillation about the second rotation axis applies a cross-coupling error to the oscillation about the first rotation axis, and wherein the cross-coupling error varies according to the Lissajous position within the Lissajous pattern; and a driver circuit including a PLL configured to adjust a drive signal that drives the oscillation about the first rotation axis. The PLL is configured to generate a PLL signal based on a phase error of the oscillation about the first rotation axis. The PLL includes a compensation circuit configured to receive the PLL signal and the Lissajous position within the Lissajous pattern, and apply a compensation value to the PLL signal to generate a compensated PLL signal for generating the drive signal based on the current Lissajous position of the oscillator structure within the Lissajous pattern.

[0144] The techniques described in this disclosure can be implemented, at least in part, in hardware, software, firmware, or any combination thereof. For example, various aspects of the described techniques can be implemented within one or more processors, including one or more microprocessors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable logic arrays (FPGAs), programmable logic controllers (PLCs), or any other equivalent integrated or discrete logic circuit system, and any combination of such components. The terms "processor" or "processing circuit system" can generally refer to any of the aforementioned logic circuit systems, alone or in combination with other logic circuit systems or any other equivalent circuit systems. Control units, including hardware, can also perform one or more of the techniques of this disclosure. Control units can use electrical signals and digital algorithms to perform their receiving, analyzing, and controlling functions, which may also include correction functions. Such hardware, software, and firmware can be implemented within the same device or in separate devices to support the various techniques described in this disclosure.

[0145] One or more aspects of this disclosure can be implemented as a non-transitory computer-readable recording medium on which a program embodying a method / algorithm is recorded, the program being used to instruct a processor to execute the method / algorithm. Therefore, the non-transitory computer-readable recording medium can have electronically readable control signals stored thereon that cooperate (or are capable of cooperating with) a programmable computer system to cause the corresponding method / algorithm to be executed. The non-transitory computer-readable recording medium can be, for example, a CD-ROM, DVD, Blu-ray disc, RAM, ROM, PROM, EPROM, EEPROM, flash memory, or electronic storage device.

[0146] Although various embodiments have been disclosed, it will be apparent to those skilled in the art that various changes and modifications can be made to achieve some of the advantages of the concepts disclosed herein without departing from the spirit and scope of the invention. It will be apparent to those skilled in the art that other components performing the same function can be suitably substituted. It should be understood that other embodiments can be utilized and structural or logical changes can be made without departing from the scope of the invention. It should be noted that features interpreted with reference to specific drawings can be combined with features in other drawings, even those not explicitly mentioned. Such modifications to the general inventive concept are intended to be covered by the appended claims and their legal equivalents.

Claims

1. An oscillator control system, comprising: An oscillator structure is configured to oscillate simultaneously around a first rotation axis and a second rotation axis according to a Lissajous pattern that is repeated frame by frame, wherein the oscillation around the second rotation axis applies a deterministic cross-coupling error to the oscillation around the first rotation axis, and wherein the deterministic cross-coupling error varies according to the Lissajous position within the Lissajous pattern. The driver circuit is configured to generate a first drive signal for driving the oscillator structure around the first rotation axis at a first drive frequency and to generate a second drive signal for driving the oscillator structure around the second rotation axis at a second drive frequency different from the first drive frequency, to generate the Lissajous pattern. The driver circuitry includes a phase-locked loop (PLL) configured to regulate the oscillation about the first rotation axis. The PLL is configured to generate a PLL signal based on the phase error of the oscillation about the first rotation axis. The PLL includes a compensation circuit configured to receive the PLL signal and a Lissajous position signal indicating the position of the Lissajous within the Lissajous pattern. The compensation circuit is configured to apply a compensation value to the PLL signal to modify the PLL signal to generate a compensated PLL signal for generating the first drive signal, wherein the compensation circuit is configured to determine the compensation value based on the Lissajous position indicated by the Lissajous position signal.

2. The oscillator control system of claim 1, wherein the compensation circuit is configured to receive compensation information including a first set of compensation values, wherein each compensation value in the first set of compensation values ​​is encoded to a different Lissajous position within the Lissajous pattern, and The compensation circuit is configured to select a compensation value from the first set of compensation values ​​based on the Lissajous position indicated by the Lissajous position signal, and apply the selected compensation value from the first set of compensation values ​​to the PLL signal, thereby modifying the PLL signal to generate a compensated PLL signal for generating the first drive signal.

3. The oscillator control system of claim 2, wherein the selected compensation value corresponds to a predetermined component of the deterministic cross-coupling error occurring at the Lissajous position indicated by the Lissajous position signal.

4. The oscillator control system of claim 2, wherein the compensation circuit is configured to use a selected compensation value to filter the deterministic cross-coupling error from the PLL signal to generate the compensated PLL signal.

5. The oscillator control system according to claim 1, wherein the deterministic cross-coupling error is a periodic cross-coupling error, the periodic cross-coupling error changing sinusoidally according to the Lissajous position within the Lissajous pattern.

6. The oscillator control system according to claim 1, further comprising: A Lissajous position counter is configured to generate the Lissajous position signal as a plurality of counter values ​​output in discrete incremental order, wherein each counter value corresponds to a different Lissajous position within the Lissajous pattern.

7. The oscillator control system of claim 6, wherein the compensation circuit is configured to receive compensation information including a first set of compensation values, wherein each compensation value in the first set of compensation values ​​is encoded to a different Lissajous position within the Lissajous pattern. The compensation circuit is configured to select a compensation value from the first set of compensation values ​​based on the Lissajous position indicated by the Lissajous position signal, and apply the selected compensation value from the first set of compensation values ​​to the PLL signal, thereby modifying the PLL signal to generate a compensated PLL signal for generating the first drive signal. The compensation circuit includes a lookup table, each counter value of which is assigned to a different compensation value in the first set of compensation values. The lookup table is configured to receive the Lissajous position signal and select the compensation value from the first set of compensation values ​​based on the counter value received in the Lissajous position signal.

8. The oscillator control system of claim 6, wherein the plurality of counter values ​​are a predetermined number of counter values, and the Lissajous position counter is configured to be reset frame by frame.

9. The oscillator control system of claim 1, wherein the PLL includes a phase error detector configured to generate a phase error signal based on a phase deviation between the oscillation and a desired oscillation phase about the first rotation axis, and to output the phase error signal as the PLL signal.

10. The oscillator control system of claim 8, wherein the deterministic cross-coupling error is superimposed on the phase error signal.

11. The oscillator control system according to claim 1, further comprising: A mirror position measurement circuit is configured to measure the rotation angle of the oscillator structure about the first rotation axis and generate a measurement signal indicating the measurement phase of the oscillation about the first rotation axis. The PLL includes a phase error detector configured to receive the measurement signal and a reference signal indicating a reference phase, generate a phase error signal based on the phase deviation between the measurement phase and the reference phase, and output the phase error signal as the PLL signal.

12. The oscillator control system of claim 11, wherein the measured phase is systematically deviated from the reference phase in a sinusoidal manner.

13. The oscillator control system according to claim 11, wherein: The mirror position measurement circuit is configured to detect the event time of the MEMS mirror based on the measured rotation angle, wherein the event time is detected at each moment when the measured rotation angle intersects with a target rotation angle among one or more target rotation angles. The reference signal includes the expected event time of the oscillator structure oscillating around the first rotation axis, and The phase error signal represents the time difference between the detected event time and the expected event time.

14. The oscillator control system according to claim 13, wherein: The detected event time is the zero-crossing time when the rotation angle of the MEMS mirror is 0° when the MEMS mirror oscillates around the first rotation axis, and The expected event time is the expected zero-crossing time when the rotation angle of the MEMS mirror is expected to be 0°.

15. The oscillator control system according to claim 1, wherein the PLL comprises: A phase error detector is configured to generate a phase error signal based on the phase deviation between the oscillation and the expected oscillation phase about the first rotation axis and to output the phase error signal. as well as A loop filter is configured to receive the phase error signal, generate a loop filter signal based on the phase error signal, and output the loop filter signal as the PLL signal, wherein the loop filter signal modulates at least one property of the first drive signal.

16. The oscillator control system according to claim 2, wherein the sum of the first set of compensation values ​​is zero.

17. The oscillator control system according to claim 2, wherein: The compensation information includes a second set of compensation values, wherein each compensation value in the second set of compensation values ​​is encoded to a different Lissajous position within the Lissajous pattern, and The compensation circuit is configured to select a compensation value from the second set of compensation values ​​based on the Lissajous position indicated by the Lissajous position signal, and apply the selected compensation value from the second set of compensation values ​​to the PLL signal, thereby modifying the PLL signal to generate the compensated PLL signal for generating the first drive signal.

18. A method for controlling an oscillator structure, the method comprising: The oscillator structure is driven to oscillate simultaneously around a first rotation axis and a second rotation axis according to a Lissajous pattern, the Lissajous pattern being repeated frame by frame, wherein the oscillation around the second rotation axis applies a deterministic cross-coupling error to the oscillation around the first rotation axis, wherein the deterministic cross-coupling error varies according to the Lissajous position within the Lissajous pattern. The oscillator structure is further driven by generating a first drive signal for driving the oscillator structure around the first rotation axis at a first drive frequency, and generating a second drive signal for driving the oscillator structure around the second rotation axis at a second drive frequency different from the first drive frequency, so as to generate the Lissajous pattern. The oscillation around the first rotation axis is regulated using a phase-locked loop (PLL), including generating a PLL signal based on the phase error of the oscillation around the first rotation axis. as well as The PLL signal is compensated using a Lissajous position signal indicating the Lissajous position within the Lissajous pattern and compensation information including a first set of compensation values, wherein each compensation value in the first set of compensation values ​​is encoded to a different Lissajous position within the Lissajous pattern. The compensation of the PLL signal further includes selecting a compensation value from the first set of compensation values ​​based on the Lissajous position indicated by the Lissajous position signal, and applying the selected compensation value from the first set of compensation values ​​to the PLL signal, thereby modifying the PLL signal to generate a compensated PLL signal for generating the first drive signal.

19. An oscillator control system, comprising: An oscillator structure is configured to oscillate simultaneously around a first rotation axis and a second rotation axis at different frequencies, wherein the oscillation around the second rotation axis applies a systematic phase error to the oscillation around the first rotation axis, and wherein the systematic phase error changes sinusoidally with time. The driver circuit is configured to generate a first drive signal for driving the oscillator structure about the first rotation axis and a second drive signal for driving the oscillator structure about the second rotation axis. The driver circuitry includes a primary phase-locked loop (PLL) configured to regulate the oscillation about the first rotation axis. The primary PLL includes a primary phase error detector configured to generate and output a phase error signal based on the phase deviation of the oscillation about the first rotation axis from the expected oscillation phase, wherein a systematic phase error is superimposed on the phase error signal. The primary PLL includes a compensation circuit configured to receive the phase error signal, copy the systematic phase error from the phase error signal, and subtract the copied systematic phase error from the phase error signal to generate a compensated phase error signal for generating the first drive signal.

20. The oscillator control system according to claim 19, wherein: The primary PLL includes a primary loop filter configured to receive the compensated phase error signal, generate a loop filter signal based on the compensated phase error signal, and output the loop filter signal that adjusts at least one property of the first drive signal. The compensation circuit is coupled to the primary phase error detector and the primary loop filter, and is also coupled between the primary phase error detector and the primary loop filter.

21. The oscillator control system according to claim 19, wherein: The compensation circuit includes: A shared voltage-controlled oscillator (VCO) with a compensation PLL and amplitude control loop, the VCO being configured to generate a VCO output signal as the replicated systematic phase error; and A subtractor is configured to subtract the VCO output signal from the phase error signal to generate the compensated phase error.

22. The oscillator control system according to claim 21, wherein: The compensation PLL is configured to receive the phase error signal and adjust the phase of the VCO output signal to synchronize with the phase of the systematic phase error. The amplitude control loop is configured to receive the phase error signal and adjust the amplitude of the VCO output signal to match the amplitude of the systematic phase error.

23. The oscillator control system according to claim 19, wherein: The compensation circuit includes: A first compensation PLL and a first amplitude control loop are shared by a first voltage-controlled oscillator (VCO), and the first VCO is configured to generate a first VCO output signal as a first part of the replicated systematic phase error; A second compensation PLL and a second amplitude control loop are shared with the second VCO, the second VCO being configured to generate a second VCO output signal as a second portion of the replicated systematic phase error; and The subtractor subtracts the first VCO output signal and the second VCO output signal from the phase error signal to generate the compensated phase error.

24. The oscillator control system according to claim 23, wherein: The first compensation PLL is configured to adjust the phase of the first VCO output signal to synchronize with the phase of the first spectral component of the systematic phase error. The first amplitude control loop is configured to adjust the amplitude of the first VCO output signal to match the amplitude of the first spectral component of the systematic phase error. The second compensation PLL is configured to adjust the phase of the second VCO output signal to synchronize with the phase of the second spectral component of the systematic phase error. The second amplitude control loop is configured to adjust the amplitude of the second VCO output signal to match the amplitude of the second spectral component of the systematic phase error.

25. A method for controlling an oscillator structure, the method comprising: The oscillator structure is driven to oscillate simultaneously around a first rotation axis and a second rotation axis at different frequencies, wherein the oscillation around the second rotation axis applies a systematic phase error to the oscillation around the first rotation axis, and wherein the systematic phase error changes sinusoidally with time. The oscillator structure is further driven by generating a first drive signal for driving the oscillator structure around the first rotation axis and generating a second drive signal for driving the oscillator structure around the second rotation axis. Adjusting the oscillation around the first rotation axis using a primary phase-locked loop (PLL) includes generating a phase error signal based on the phase deviation between the oscillation around the first rotation axis and the expected oscillation phase, and outputting the phase error signal, wherein the systematic phase error is superimposed on the phase error signal; as well as Compensating the phase error signal includes copying the systematic phase error from the phase error signal and subtracting the copied systematic phase error from the phase error signal to generate a compensated phase error signal for generating the first drive signal.

Images