A solid-state laser radar scanning system based on a micro-mirror array

Abstract

The application relates to the technical field of laser radars, and discloses a solid-state laser radar scanning system based on a micro-mirror array, which comprises a micro-mirror array scanning unit, a transmitting optical unit, a receiving optical unit, a driving control unit and a signal processing unit. By adopting an N*M distributed MEMS micro-mirror array, each micro-mirror is configured with independent electrostatic driving electrodes and angle detection electrodes, full-view field splicing is completed through a scanning view field splicing sub-module, the scanning view field is expanded without increasing the size of a single mirror, and the high resonance frequency of the micro-mirror is ensured; the driving control unit adopts a FPGA and a multi-channel DAC chip cascade architecture, can correct scanning distortion in real time according to echo information, and improves scanning linearity; an arrayed diaphragm corresponding to the micro-mirror array is arranged in the transmitting optical unit, can suppress ghost images and crosstalk in multi-beam transmission, and improves imaging quality and detection precision.

Description

Technical Field

[0001] This invention relates to the field of lidar technology, and more specifically, to a solid-state lidar scanning system based on a micro-mirror array. Background Technology

[0002] With the rapid development of autonomous driving, intelligent robots and other fields, LiDAR has become a core sensor for environmental perception systems due to its advantages such as high detection accuracy, fast response speed and strong anti-interference ability.

[0003] Among numerous technological approaches, hybrid solid-state lidar based on MEMS (Micro-Electro-Mechanical Systems) micromirrors is considered one of the most promising solutions for automotive lidar applications due to its advantages such as small size, low power consumption, and controllable cost. However, existing MEMS lidar technology still faces the following bottlenecks: First, there is a conflict between scanning field of view and frame rate. Existing solutions mostly use a single MEMS micromirror. Due to limitations in physical size and driving capability, the mechanical deflection angle of a single mirror is limited, resulting in a small scanning field of view. Although the field of view can be expanded by increasing the mirror size, this will significantly increase the rotational inertia of the mirror, causing its resonant frequency to drop sharply, which cannot meet the high frame rate (e.g., ≥20fps) requirements of automotive applications.

[0004] Second, poor scanning linearity and optical path crosstalk. Most MEMS micromirrors use open-loop control, and their deflection angle has a non-linear relationship with the driving voltage, resulting in poor scanning linearity and affecting point cloud accuracy. In addition, in multi-beam emission schemes, when multiple beams share the same set of optical lenses, the lack of effective physical isolation design easily leads to reflections within the lens group, forming "ghost images" and inter-channel crosstalk, which seriously reduces imaging quality and detection accuracy.

[0005] Third, limitations of ranging systems. Frequency modulated continuous wave (FMCW) ranging has advantages such as strong anti-interference capability and simultaneous velocity measurement, but it has a blind zone at close range. While time-of-flight (ToF) ranging has no blind zone, it performs poorly in terms of accuracy and anti-interference at long distances. A simple "dual-mode coexistence" scheme leads to system complexity, clock asynchrony, and difficulty in achieving seamless ranging with high accuracy across the entire range.

[0006] Therefore, developing a solid-state lidar system that can simultaneously achieve a large field of view, high frame rate, high linearity, high ranging accuracy, and strong anti-interference capability is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0007] The present invention aims to provide a solid-state lidar scanning system based on a micro-mirror array to solve the problems in the prior art, such as the inability to balance scanning field of view and scanning frequency, poor scanning linearity, severe optical crosstalk, and the inability to guarantee full-range ranging accuracy.

[0008] To achieve the objectives of this invention, the technical solution is as follows: A solid-state lidar scanning system based on a micro-mirror array includes a micro-mirror array scanning unit, a transmitting optical unit, a receiving optical unit, a drive control unit, and a signal processing unit.

[0009] The micromirror array scanning unit employs an N×M distributed MEMS micromirror array. This array comprises an X-axis resonant scanning subarray and a Y-axis quasi-static scanning subarray. The X-axis resonant scanning subarray consists of at least two first MEMS micromirrors, and the Y-axis quasi-static scanning subarray consists of at least two second MEMS micromirrors. The first and second MEMS micromirrors are arranged in a rectangular symmetrical configuration. Each MEMS micromirror is equipped with an independent electrostatic drive electrode and an angle detection electrode.

[0010] The transmitting optical unit adopts a coaxial polarization beam splitter optical path structure, with a narrow linewidth fiber laser, collimating lens group, polarization beam splitter prism, quarter-wave plate, micro-mirror array scanning unit, off-axis parabolic mirror array, and wide-angle scanning lens group arranged sequentially along the beam transmission direction. The off-axis parabolic mirror array contains off-axis parabolic mirrors with the same number as the MEMS micro-mirrors, and each off-axis parabolic mirror corresponds one-to-one with each MEMS micro-mirror. The wide-angle scanning lens group is a zoom optical system composed of multiple positive and negative lenses. In particular, the transmitting optical unit also has an array of apertures corresponding one-to-one with each MEMS micro-mirror to suppress ghosting and crosstalk during multi-beam transmission.

[0011] The receiving optical unit is arranged sequentially along the beam transmission direction, consisting of a wide-angle receiving mirror group and an array of photodetectors. The wide-angle receiving mirror group adopts an image-side telecentric optical path structure, and its receiving field of view matches the scanning field of view of the transmitting optical unit. The array of photodetectors uses an APD detector array with the same number of scanning fields of view as the MEMS micromirror array, and the receiving channel of each APD detector corresponds one-to-one with the scanning field of view of each MEMS micromirror.

[0012] The drive control unit adopts an FPGA and multi-channel DAC chip cascade architecture. The FPGA integrates a mirror array synchronization drive module, which includes a phase-locked loop submodule, a drive voltage pre-distortion correction submodule, and a scanning field-of-view stitching submodule. The phase-locked loop submodule outputs multiple in-phase and frequency-coordinated resonant drive signals to the X-axis resonant scanning subarray. The drive voltage pre-distortion correction submodule outputs corrected drive voltages to each MEMS micromirror based on a polynomial pre-distortion model. The scanning field-of-view stitching submodule performs physical timing synchronization and sub-pixel-level coordinate matching of the scanning trajectories of each MEMS micromirror, achieving full-field-of-view stitching without blind spots.

[0013] The signal processing unit is electrically connected to each output channel of the arrayed photodetector. It adopts a dual ranging system with FMCW coherent detection as the primary method and TDC pulse ToF ranging as the secondary method, and incorporates a clock division module and a fusion time-frequency imaging algorithm module. The clock division module uses the laser-modulated clock of the FMCW coherent detection as the reference clock for TDC time measurement, achieving clock homogeneity and timing synchronization in the dual ranging system.

[0014] Furthermore, the signal processing unit forms a multi-path closed-loop feedback link with the drive control unit. The signal processing unit extracts scanning distortion information, phase synchronization error information, and field-of-view stitching deviation information based on the echo signal, and feeds them back in real time to the drive voltage pre-distortion correction submodule, the phase-locked loop submodule, and the scanning field-of-view stitching submodule, respectively, to dynamically update the operating parameters of the corresponding modules.

[0015] As a preferred technical solution, the MEMS micro-mirror array is a 2×2 distributed four-unit MEMS micro-mirror array, wherein two first MEMS micro-mirrors constitute an X-axis resonant scanning sub-array, and two second MEMS micro-mirrors constitute a Y-axis quasi-static scanning sub-array.

[0016] As a preferred technical solution, the driving voltage predistortion correction submodule adopts a third-order polynomial predistortion model.

[0017] As a preferred technical solution, the execution steps of the fusion time-frequency imaging algorithm module include: range-dimensional FFT, Doppler-dimensional FFT, STFT time-frequency analysis, and DWT wavelet denoising and reconstruction.

[0018] Compared with the prior art, the solid-state lidar scanning system based on a micro-mirror array provided by the present invention has the following advantages: Achieving a balance between a large field of view and a high frame rate: By employing an N×M distributed MEMS micromirror array, which includes an X-axis resonant scanning subarray and a Y-axis quasi-static scanning subarray, each micromirror is equipped with an independent electrostatic driving electrode and an angle detection electrode. The scanning field of view stitching submodule completes the full field of view stitching without blind spots, expanding the scanning field of view without increasing the size of a single mirror, while ensuring the high resonant frequency of the micromirrors, thus meeting the requirements of a large field of view and high frame rate in automotive applications.

[0019] Improved scanning linearity and imaging quality: The drive control unit adopts an FPGA and multi-channel DAC chip cascade architecture, with a built-in mirror array synchronous drive module, forming a multi-path closed-loop feedback link with the signal processing unit. It can correct scanning distortion in real time according to the echo information to improve scanning linearity. The transmitting optical unit is equipped with arrayed apertures that correspond one-to-one with the micro mirror array, which can suppress ghost images and crosstalk in multi-beam transmission, and improve imaging quality and detection accuracy.

[0020] Achieving full-range, high-precision ranging capability: This system innovatively employs a dual-mode heterogeneous ranging architecture, primarily using FMCW and secondarily using ToF, achieving clock synchronization and seamless range switching through clock division and FPGA timing control. This approach combines the advantages of FMCW's long-range high precision and anti-interference capabilities with ToF's lack of near-range blind spots, covering a ranging range from 0.5m to 200m. Furthermore, the proposed "FFT-STFT-DWT" fusion time-frequency imaging algorithm effectively extracts target signals against strong noise backgrounds, further ensuring high-precision detection across the entire range. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a block diagram of the overall architecture of the solid-state lidar scanning system described in this invention; Figure 2 This is a flowchart of the execution steps of the fusion time-frequency imaging algorithm module described in this invention; Figure 3 This is a block diagram of the multi-path closed-loop feedback architecture between the drive control unit and the signal processing unit described in this invention; Figure 4 This is a flowchart of the clock co-source and timing coordination control of the dual ranging system described in this invention. Detailed Implementation

[0023] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention. It should be noted that relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0024] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings and embodiments.

[0025] Example 1: This embodiment mainly illustrates the micro-mirror array arrangement structure and the specific implementation method of the transmitting / receiving optical path of the system described in this invention, aiming to solve the problems of the inability to balance the scanning field of view and frequency and optical path crosstalk.

[0026] like Figure 1 As shown, a solid-state lidar scanning system based on a micro-mirror array includes a micro-mirror array scanning unit, a transmitting optical unit, a receiving optical unit, a drive control unit, and a signal processing unit that are electrically connected to each other.

[0027] The micromirror array scanning unit in this embodiment is a 2×2 distributed four-element MEMS micromirror array. This array includes an X-axis resonant scanning subarray and a Y-axis quasi-static scanning subarray.

[0028] The X-axis resonant scanning subarray consists of two first MEMS micromirrors, namely the first micromirror M1 and the second micromirror M2. The first micromirror M1 and the second micromirror M2 are arranged along the X-axis to achieve fast resonant scanning in the horizontal direction. The resonant frequency is 427Hz and the maximum mechanical deflection angle in the X-axis direction is ±5.2°.

[0029] The Y-axis quasi-static scanning subarray consists of two second MEMS micromirrors, namely the third micromirror M3 and the fourth micromirror M4. The third micromirror M3 and the fourth micromirror M4 are arranged along the Y-axis to achieve quasi-static scanning in the vertical direction. Their operating frequency is 1kHz and their maximum mechanical deflection angle in the Y-axis direction is ±5.2°.

[0030] In a preferred embodiment of the present invention, the first micromirror M1, the second micromirror M2, the third micromirror M3, and the fourth micromirror M4 are arranged in a rectangular symmetrical configuration. M1 and M2 are arranged along the X-axis, and M3 and M4 are arranged along the Y-axis. The center-to-center distance between adjacent micromirrors is 4.2 mm. The aperture of each of the first micromirror M1, the second micromirror M2, the third micromirror M3, and the fourth micromirror M4 is 4.2 mm. Each MEMS micromirror (i.e., M1-M4) is equipped with an independent electrostatic drive electrode and an angle detection electrode. The angle detection electrode collects the actual deflection angle of the micromirror in real time and feeds it back to the drive control unit.

[0031] The transmitting optical unit adopts a coaxial polarization beam splitting optical path structure, and is arranged in sequence along the beam transmission direction, including a 1550nm narrow linewidth fiber laser, an aspherical collimating lens group, a polarization beam splitting prism, a quarter wave plate, a micro-mirror array scanning unit, an off-axis parabolic mirror array, an arrayed aperture, and a wide-angle scanning lens group.

[0032] (1) A 1550nm narrow-linewidth fiber laser, for example, adopts a master oscillator plus power amplifier (MOPA) structure. It can realize time-division switching between linear frequency modulated continuous wave (FMCW) mode and pulse mode through FPGA timing control module. The two modes share the same laser resonant cavity and output optical path; the output laser pulse width is 10ns, the repetition frequency is 100kHz, the peak power is 1000W, and the beam... The factor is ≤1.2, and the output port uses SMF-28e standard single-mode fiber with a numerical aperture of 0.122.

[0033] (2) An aspherical collimating lens group, for example, consists of a spherical lens and an even-order aspherical lens, used to collimate the laser. The front surface curvature radius of the spherical lens is 76.508 mm, the rear surface curvature radius is -32.364 mm, the lens thickness is 5 mm, the glass material is SF18, the effective focal length is 76.47 mm, and the diameter of the collimated Gaussian beam at the MEMS micro-mirror mirror is 2.9 mm, which can be completely projected onto the 4.2 mm diameter micro-mirror mirror with an energy loss of less than 25%.

[0034] (3) The polarizing beam splitter is used in conjunction with the quarter-wave plate to achieve coaxial beam splitting and isolation of the emitted light and the echo light, with an isolation of ≥40dB, to avoid the emitted light from directly interfering with the receiving channel.

[0035] (4) An off-axis parabolic mirror array, for example, includes four off-axis parabolic mirrors. The four off-axis parabolic mirrors are set one-to-one with M1, M2, M3 and M4. The effective focal length of each off-axis parabolic mirror is 101.6mm, the parent focal length is 50.8mm, the aperture is 25mm, the offset angle is 30°, and the Y-axis offset is 27.6mm. The MEMS micromirror is placed at the focal point of the corresponding off-axis parabolic mirror. The laser beam reflected by the micromirror is incident on the off-axis parabolic mirror and forms a parallel beam after reflection. The distance between the parallel beam and the principal optical axis is linearly related to the scanning angle of the micromirror, thus achieving linearized scanning. An array of apertures is positioned between the off-axis parabolic mirror array and the wide-angle scanning lens group, corresponding one-to-one with the four micro-mirrors M1-M4. The aperture of each aperture matches the exit beam aperture of the corresponding off-axis parabolic mirror. For example, the aperture is 26mm. A light-absorbing barrier structure is set between adjacent apertures. The inner wall of the aperture is coated with an anti-light coating, which suppresses stray light of non-target beams by ≥60dB. This effectively blocks ghost images formed by reflections of beams from adjacent channels within the lens group, avoiding crosstalk between multiple beams.

[0036] (5) The wide-angle scanning lens group consists of 14 spherical lenses, which are used to achieve large field of view scanning and f-theta distortion suppression; its effective focal length is 14.05 mm, the f-theta distortion of the whole field of view is ≤0.43%, and the image plane offset of the scanning beam is linearly positively correlated with the deflection angle of the MEMS micro-mirror; among them, the 14 spherical lenses are arranged sequentially from the first lens to the fourteenth lens along the beam transmission direction, and the specific parameters are shown in Table 1: Table 1 ; The receiving optical unit includes a wide-angle receiving mirror group and an arrayed photodetector arranged sequentially along the beam transmission direction; the light inlet of the receiving optical unit corresponds to the reflected light output end of the polarization beam splitter of the transmitting optical unit.

[0037] The wide-angle receiving lens group adopts an image-side telecentric optical path structure, and its receiving field of view is matched with the scanning field of view of the transmitting end. In an optional embodiment of the present invention, the entrance pupil diameter of the wide-angle receiving lens group is 3mm, the F-number is 0.55, and the receiving field of view is 38.76°×59.81°, which perfectly matches the scanning field of view of the transmitting end. The wide-angle receiving lens group consists of a front meniscus negative lens and a rear group of three positive lenses. The front meniscus negative lens has a radius of curvature of -86.535mm, a thickness of 2.210mm, and is made of LAF2 glass. The rear group of three positive lenses are, in sequence, a first positive lens, a second positive lens, and a third positive lens. The first positive lens has a radius of curvature of 20.361mm, a thickness of 11.049mm, and is made of PSK3 glass. The second positive lens has a radius of curvature of 38.689mm, a thickness of 6.147mm, and is made of SF3 glass. The third positive lens has a radius of curvature of 7.452mm, a thickness of 15.799mm, and is made of K7 glass. In this embodiment, the wide-angle receiving lens group has a full field-of-view relative illumination of ≥95%, optical distortion of ≤5%, and a back focal length of 1.5mm, which can efficiently focus the echo beam of the large field of view onto the photosensitive surface of the arrayed photodetector.

[0038] For example, an arrayed photodetector uses a 1×4 InGaAs APD detector array, adapted to the 1550nm operating band. The four InGaAs APD detectors are integrated on the same common substrate chip. The photosensitive surface diameter of a single InGaAs APD detector is 1mm. The receiving channels of the four InGaAs APD detectors correspond one-to-one with the four scanning fields of M1, M2, M3, and M4. The echo beam is received using a space-division multiplexing method. Each InGaAs APD detector output channel is equipped with an independent T-network transimpedance amplifier circuit. The T-network transimpedance amplifier circuit uses an OPA657 operational amplifier, and the feedback network consists of a first resistor. Second resistor Third resistor composition, The value is 100kΩ. The value is 9kΩ. With a value of 1kΩ, the circuit has a transimpedance gain of 1MΩ and a bandwidth of 200MHz, enabling high-gain, low-noise amplification of weak echo current signals in the nA range.

[0039] The drive control unit adopts an FPGA and four-channel DAC chip cascade architecture. The FPGA uses AlteraCyclone IV series chips, and the four-channel DAC chip uses AD5764 chips with 16-bit resolution. The four output channels are respectively connected to the electrostatic drive electrodes of the four MEMS micro mirrors to realize the independent drive of the four micro mirrors.

[0040] The signal processing unit is electrically connected to the InGaAs APD detector array and the drive control unit, forming a closed-loop feedback.

[0041] Example 2 This embodiment will focus on illustrating the closed-loop feedback mechanism between the drive control unit and the signal processing unit of the present invention, as well as the specific control logic for seamless stitching of the multi-micromirror field of view, specifically including: The FPGA of the drive control unit has a built-in galvanometer array synchronous drive module, which includes a phase-locked loop submodule, a drive voltage pre-distortion correction submodule, and a scanning field of view stitching submodule. At the same time, the drive control unit establishes a multi-path closed-loop feedback link with the signal processing unit, receives scanning distortion information, phase synchronization error information, and field of view stitching deviation information returned by the signal processing unit, and dynamically adjusts the operating parameters of each submodule.

[0042] The phase-locked loop submodule is used to drive the two first MEMS micromirrors M1 and M2 of the X-axis resonant scanning subarray. The submodule has a built-in digital phase-locked loop and outputs two resonant drive signals with the same frequency and phase to the electrostatic drive electrodes of M1 and M2. The phase error of the two resonant drive signals is ≤0.1°, which ensures the synchronous resonant scanning of M1 and M2 and avoids misalignment in the field of view stitching. At the same time, the submodule can adjust the phase difference of the two drive signals in real time according to the phase synchronization error information fed back by the signal processing unit, so as to ensure the synchronization stability of long-term operation.

[0043] The driving voltage predistortion correction submodule performs nonlinear correction on the driving voltage of the MEMS micromirror based on a third-order polynomial predistortion model. The expression of the third-order polynomial predistortion model is as follows: ; in The target mechanical deflection angle of the MEMS micromirror. The corrected driving voltage, These are third-order coefficients, taking values ​​of , These are second-order coefficients, taking values ​​of , These are first-order coefficients, taking values ​​of , For constant terms, the value is... ; This model pre-distorts the tilt-voltage nonlinearity of the MEMS micromirror, and the scanning linearity of the MEMS micromirror after correction is ≥99.8%, ensuring precise control of the scanning angle. At the same time, the third-order coefficients, second-order coefficients, first-order coefficients and constant terms of the model can be updated in real time according to the echo nonlinear phase error returned by the signal processing unit, realizing adaptive correction.

[0044] The scanning field-of-view stitching submodule adopts a hybrid mechanism of spatial division multiplexing and time division multiplexing to perform physical time synchronization and sub-pixel level coordinate matching of the scanning trajectories of the four MEMS micro-mirrors, achieving blind-zone-free stitching across the entire field of view. The specific execution steps are as follows: Step 1: First, establish a global scanning coordinate system, map the local scanning coordinate systems of the four MEMS micro mirrors to the global scanning coordinate system, and obtain the coordinate mapping matrix of the scanning trajectory of each micro mirror; Step 2: Then, a physical overlap area is set at the edge of the scanning field of view of adjacent micromirrors. In this embodiment, the far-field divergence angle of the beam emitted by a single micromirror is 0.067°, and the angle of the overlap area is set to 0.1°, which is 1.5 times the divergence angle. Through FPGA timing control, the scanning beams of adjacent micromirrors scan the overlap area in a time-division manner in a very short time. The scanning spot sequence forms a continuous coverage overlap area without physical gaps in the far field, realizing seamless connection of the entire field of view, rather than physical superposition and fusion of beams in space. Bilinear interpolation coordinate matching is performed on the scanning points in the overlap area to eliminate splicing misalignment. Step 4: Finally, the scanning trajectories of the four micromirrors are synchronized in time, with a synchronization error of ≤1μs. The full field of view scanning trajectory is stitched together, with a stitching blind zone of ≤0.05°. The stitched full field of view is 38.76°×59.81°. Simultaneously, this submodule can update the coordinate mapping matrix in real time based on the field-of-view stitching deviation information fed back by the signal processing unit, thereby correcting the stitching error. Specifically, such as... Figure 3 As shown, the signal processing unit and the drive control unit form a multi-path closed-loop feedback link. The signal processing unit extracts the following from the echo signal: The scanning distortion information (such as nonlinear phase error) is fed back to the drive voltage predistortion correction submodule for dynamically updating the polynomial coefficients a, b, c, d.

[0045] Phase synchronization error information is fed back to the phase-locked loop submodule for fine-tuning the phase of the drive signal to ensure long-term synchronization.

[0046] Field-of-view stitching deviation information (such as sudden changes in light intensity or coordinate misalignment at the stitching edge) is fed back to the scanning field-of-view stitching submodule to update the coordinate mapping matrix and correct stitching errors.

[0047] Through this end-to-end closed-loop adaptive correction, the system can operate stably in its optimal state for a long period of time.

[0048] Example 3 This embodiment mainly illustrates how the present invention achieves high-precision target detection across the entire range by integrating two ranging systems, FMCW and TDC, and by employing advanced time-frequency analysis algorithms.

[0049] In this embodiment, as Figure 4 As shown, the signal processing unit is electrically connected to the four output channels of the arrayed photodetector. It adopts a dual ranging system with FMCW coherent detection as the main method and TDC pulse ToF ranging as the auxiliary method. It has a built-in clock division module, a 2D-FFT and STFT and DWT fusion time-frequency imaging algorithm module, and a TDC time measurement module. At the same time, it forms a multi-path closed-loop feedback link with the drive control unit.

[0050] The clock divider module receives the modulated clock output from the 1550nm narrow-linewidth fiber laser. After frequency division, it serves as the reference clock for the TDC time measurement module, achieving clock homogeneity for the dual ranging system and avoiding timing deviations and measurement asynchrony issues caused by independent clocks. Simultaneously, through the timing control module built into the FPGA, seamless switching between the FMCW and TDC ranging modes is achieved. The TDC mode operates in the 0.5m-8m short-range range, while the FMCW mode operates in the 5m-200m medium-to-long-range range. The overlapping range uses a weighted fusion algorithm to output the final ranging result, avoiding range conflicts.

[0051] The dual-mode time-division switching logic of the laser is as follows: The FPGA timing control module switches the laser's operating mode according to the scanning line cycle. Within a single scanning line cycle, the first 10% of the time slots control the laser to operate in pulse mode, completing full field-of-view scanning and ToF ranging in the near-field area. The latter 90% of the time slot control laser operates in linear frequency modulated continuous wave mode to complete full-field FMCW coherent detection. The scanning trajectories of the two modes completely overlap, ensuring the spatial consistency of the ranging results.

[0052] The TDC time measurement module uses the TDC7200 chip. The START port of the TDC7200 chip is connected to the synchronous trigger signal of the 1550nm narrow linewidth fiber laser, and the STOP port is connected to the echo time discrimination signal output by the arrayed photodetector. The TDC7200 chip has a time measurement resolution of 55ps and a measurement standard deviation of ≤120ps within a 200ns range. It is used for high-precision pulse ToF ranging in the near-range blind zone, solving the problem of near-range ranging blind zone in the FMCW system.

[0053] The fusion time-frequency imaging algorithm module is used for high-precision extraction of target distance and velocity information under FMCW system. The execution steps are as follows: Figure 2 As shown, it specifically includes: Step 1: Mix the linear frequency modulated continuous wave output from the 1550nm narrow linewidth fiber laser with the echo signal received by the arrayed photodetector. After low-pass filtering with a cutoff frequency of 30MHz, a clean beat frequency signal is obtained. Perform a 1024-point distance dimension FFT operation on the beat frequency signal to extract the signal amplitude and phase information corresponding to each distance gate and generate a distance dimension spectrum image. Step 2: Stack the range-dimensional spectrum images corresponding to 128 consecutive cycles of linear frequency modulated continuous wave along the slow time dimension, perform 128-point Doppler FFT operation on the signal sequence corresponding to each range gate to generate a 2D range-Doppler image, and initially extract the initial range and velocity values ​​of the target. Step 3: Apply a Hanning window to the single-cycle beat frequency signal and perform a sliding window STFT transformation. The sliding window length is 256 points and the sliding step size is 64 points to generate a time-frequency image with a time-frequency resolution of 64 points × 16 frames. Extract the instantaneous frequency change curve of the target from the time-frequency image, correct the initial value of the target's velocity, and solve the problem of decreased velocity resolution caused by the Doppler frequency shift of high-speed moving targets. Step 4: Perform db4 wavelet 5-level DWT multi-scale decomposition on the time-frequency image generated by STFT transform to obtain 1 low-frequency approximation coefficient and 5 high-frequency detail coefficients. Use a soft thresholding function to denoise the high-frequency detail coefficients to filter out environmental noise and circuit noise. Perform inverse wavelet transform on the denoised coefficients to reconstruct the target feature time-frequency image. Extract the final distance information and final velocity information of the target from the reconstructed feature time-frequency image.

[0054] Meanwhile, the signal processing unit extracts scanning distortion information such as scanning linearity deviation and nonlinear phase error corresponding to optical distortion based on the spot characteristics of the reconstructed characteristic time-frequency image and echo signal, and feeds it back to the drive voltage pre-distortion correction submodule in real time; it extracts the phase synchronization error information of the two resonant drive signals and feeds it back to the phase-locked loop submodule in real time; it extracts the light intensity change and coordinate misalignment information at the field of view stitching and feeds it back to the scanning field of view stitching submodule in real time, realizing adaptive closed-loop correction of the entire scanning link and improving the long-term working stability and environmental adaptability of the system.

[0055] In this embodiment, the overall optical path length of the system is 10cm, and the system width and height are both no more than 5cm, resulting in a compact structure. The spot diameter at a detection distance of 100m is 17cm, the maximum spot diameter within the entire field of view is no more than 20cm, and the minimum angular resolution is 0.1°. The ranging range is from 0.5m to 200m, with a ranging accuracy better than 2ns within 8m and a ranging accuracy ≤2cm at 200m. The scanning frame rate is ≥20fps, which can meet the real-time and high-precision detection requirements of vehicle-mounted autonomous driving scenarios.

[0056] In summary, this solid-state lidar scanning system based on micro-mirror arrays employs an N×M distributed MEMS micro-mirror array, which includes an X-axis resonant scanning sub-array and a Y-axis quasi-static scanning sub-array. Each micro-mirror is equipped with an independent electrostatic driving electrode and an angle detection electrode. The scanning field of view stitching sub-module completes full-field-of-view stitching without blind spots, expanding the scanning field of view without increasing the size of a single mirror, while ensuring the high resonant frequency of the micro-mirrors, thus meeting the requirements of a large field of view and high frame rate in automotive applications.

[0057] Furthermore, this solid-state lidar scanning system based on micro-mirror arrays adopts an FPGA and multi-channel DAC chip cascade architecture for its drive control unit. It has a built-in mirror array synchronous drive module, which forms a multi-path closed-loop feedback link with the signal processing unit. It can correct scanning distortion in real time based on echo information to improve scanning linearity. The transmitting optical unit is equipped with arrayed apertures that correspond one-to-one with the micro-mirror array, which can suppress ghosting and crosstalk in multi-beam transmission and improve imaging quality and detection accuracy.

[0058] Furthermore, this solid-state lidar scanning system based on a micro-mirror array employs a dual ranging system in its signal processing unit, which primarily uses FMCW coherent detection and secondarily uses TDC pulse ToF ranging. A clock division module ensures clock homogeneity and timing synchronization between the two systems. Both modes share the same laser resonant cavity and output optical path, forming a full-range complementary system through seamless range switching. Combined with a fusion time-frequency imaging algorithm, it achieves high-precision ranging across the entire range, solving the problems of existing MEMS lidar systems where the scanning field of view and scanning frequency cannot be simultaneously optimized, scanning linearity is poor, optical path crosstalk is severe, and full-range ranging accuracy cannot be guaranteed.

[0059] The relevant modules involved in this system are all hardware system modules or functional modules that combine computer software programs or protocols with hardware in the prior art. The computer software programs or protocols involved in these functional modules are technologies known to those skilled in the art and are not improvements to this system. The improvement of this system lies in the interaction or connection between the modules, that is, in improving the overall structure of the system to solve the corresponding technical problems that this system aims to address.

[0060] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A solid-state lidar scanning system based on a micro-mirror array, characterized in that, It includes a micro-mirror array scanning unit, a transmitting optical unit, a receiving optical unit, and a drive control unit that are electrically connected to each other; The micromirror array scanning unit adopts an N×M distributed MEMS micromirror array. The MEMS micromirror array includes an X-axis resonant scanning subarray and a Y-axis quasi-static scanning subarray. The X-axis resonant scanning subarray consists of at least two first MEMS micromirrors, and the Y-axis quasi-static scanning subarray consists of at least two second MEMS micromirrors. The first and second MEMS micromirrors are arranged in a rectangular symmetrical pattern. Each first and second MEMS micromirror is equipped with an independent electrostatic driving electrode and an angle detection electrode. The transmitting optical unit adopts a coaxial polarization beam splitter optical path structure, and a narrow linewidth fiber laser, a collimating lens group, a polarization beam splitter prism, a quarter wave plate, a micro-mirror array scanning unit, an off-axis parabolic mirror array, and a wide-angle scanning lens group are arranged sequentially along the beam transmission direction. The transmitting optical unit is also equipped with an array of apertures corresponding to each MEMS micro-mirror to suppress ghosting and crosstalk during multi-beam transmission. The receiving optical unit is arranged with a wide-angle receiving mirror group and an arrayed photodetector in sequence along the beam transmission direction. The receiving field of view of the wide-angle receiving mirror group matches the scanning field of view of the transmitting optical unit. Each detection channel in the arrayed photodetector corresponds one-to-one with the scanning field of view of each MEMS micro-mirror. The drive control unit adopts an FPGA and multi-channel DAC chip cascade architecture. The FPGA has a built-in mirror array synchronous drive module, which includes a phase-locked loop submodule, a drive voltage pre-distortion correction submodule, and a scanning field-of-view stitching submodule. The phase-locked loop submodule is used to output multiple resonant drive signals with the same frequency and phase to the X-axis resonant scanning subarray. The drive voltage pre-distortion correction submodule outputs the corrected drive voltage to each MEMS micromirror based on a polynomial pre-distortion model. The scanning field-of-view stitching submodule is used to perform physical timing synchronization and sub-pixel level coordinate matching of the scanning trajectory of each MEMS micromirror to achieve full field-of-view stitching without blind spots.

2. The system according to claim 1, characterized in that, It also includes a signal processing unit, which is electrically connected to each output channel of the arrayed photodetector. The signal processing unit adopts a dual ranging system with FMCW coherent detection as the main method and TDC pulse ToF ranging as the auxiliary method, and has a built-in clock division module and a fusion time-frequency imaging algorithm module. The clock division module uses the laser-modulated clock of FMCW coherent detection as the reference clock for TDC time measurement, so as to achieve clock homogeneity and timing synchronization of the dual ranging system. The signal processing unit also forms a multi-path closed-loop feedback link with the drive control unit, which is used to extract scanning distortion information, phase synchronization error information and field of view stitching deviation information based on the echo signal, and feeds them back to the drive voltage pre-distortion correction submodule, phase locking loop submodule and scanning field of view stitching submodule in real time, so as to dynamically update the operating parameters of the corresponding modules.

3. The system according to claim 2, characterized in that, The MEMS micromirror array is a 2×2 distributed four-unit MEMS micromirror array, wherein two first MEMS micromirrors constitute the X-axis resonant scanning subarray, and two second MEMS micromirrors constitute the Y-axis quasi-static scanning subarray; the center-to-center distance between adjacent micromirrors is 4.2 mm, and the aperture of each micromirror is 4.2 mm; the maximum mechanical deflection angle of each first MEMS micromirror in the X-axis direction is ±5.2°, and the resonant frequency is 427 Hz; the maximum mechanical deflection angle of each second MEMS micromirror in the Y-axis direction is ±5.2°, and the quasi-static operating frequency is 1 kHz; The phase error of the two resonant drive signals output by the phase-locked loop submodule is ≤0.1°; The polynomial predistortion model of the driving voltage predistortion correction submodule is a third-order polynomial model, and its expression is: U(θ)=aθ³+bθ²+cθ+d, where θ is the target mechanical deflection angle and U(θ) is the corrected driving voltage; and the third-order coefficients a, second-order coefficients b, first-order coefficients c and constant term d are updated in real time according to the nonlinear phase error of the echo signal.

4. The system according to claim 2, characterized in that, The narrow-linewidth fiber laser is a 1550nm band fiber laser, employing a master oscillator plus power amplifier structure, and can achieve time-division switching between linear frequency modulated continuous wave mode and pulse mode through an FPGA timing control module; the collimating lens group consists of a spherical lens and an even-order aspherical lens; in the off-axis parabolic mirror array, each off-axis parabolic mirror is set in a one-to-one correspondence with a corresponding MEMS micro-mirror; the wide-angle scanning lens group is a zoom optical system composed of multiple positive and negative lenses, used to achieve a large field of view expansion of the scanning beam and f-theta distortion suppression.

5. The system according to claim 2, characterized in that, The wide-angle receiving lens group adopts an image-side telecentric optical path structure, with a full field of view relative illumination ≥95% and optical distortion ≤5%; the arrayed photodetector adopts a 1×4 InGaAs APD detector array, and the output channel of each APD detector is equipped with an independent T-type network transimpedance amplifier circuit.

6. The system according to claim 2, characterized in that, The specific steps for stitching together the scanning trajectories of each MEMS micromirror by the scanning field stitching submodule include: Step 1: Establish a global scanning coordinate system, map the local scanning coordinate system of each MEMS micromirror to the global scanning coordinate system, and obtain the coordinate mapping matrix of the scanning trajectory of each micromirror; Step 2: Set a physical overlap area of ​​preset width at the edge of the scanning field of view of adjacent micromirrors, so that the scanning spot sequence of adjacent micromirrors forms a continuous overlapping area without physical gaps in the far field, and perform bilinear interpolation coordinate matching on the scanning points in the overlapping area. Step 3: Synchronize the scanning trajectories of each MEMS micromirror in time, with a synchronization error ≤1μs, and complete the stitching of the full field of view scanning trajectory, with a stitching blind zone ≤0.05°.

7. The system according to claim 2, characterized in that, The specific execution steps of the fusion time-frequency imaging algorithm module are as follows: Step 1: Mix and filter the linear frequency modulated continuous wave and the echo signal to obtain the beat frequency signal. Perform a distance-dimensional FFT operation on the beat frequency signal to generate a distance-dimensional spectrum image. Step 2: Stack multiple consecutive periodic range-dimensional spectral images along the slow time dimension, perform Doppler-dimensional FFT operation on the signal sequence corresponding to each range gate to generate a 2D range-Doppler image, and extract the initial range and initial velocity values ​​of the target. Step 3: Perform a sliding window STFT transformation on the single-cycle beat frequency signal to generate a time-frequency image, extract the instantaneous frequency change curve of the target, and correct the initial value of the target's velocity. Step 4: Perform DWT multi-scale decomposition on the time-frequency image, denoise the high-frequency detail coefficients obtained from the decomposition, and then perform inverse wavelet transform to reconstruct the target feature time-frequency image, from which the final distance information and final velocity information of the target are extracted.

8. The system according to claim 2, characterized in that, The TDC time measurement module is used for pulse ToF ranging of close targets at a distance of 0.5m-8m, and the FMCW coherent detection system is used for ranging and velocity measurement of medium and long-range targets at a distance of 5m-200m. The two are connected by an FPGA timing control module to achieve seamless range switching.

9. The system according to claim 2, characterized in that, The arrayed aperture is disposed between the off-axis parabolic mirror array and the wide-angle scanning lens group. The aperture of each aperture is matched with the aperture of the emitted beam of the corresponding off-axis parabolic mirror. A light-absorbing barrier structure is provided between adjacent apertures, and the inner wall of the aperture is coated with an anti-reflective coating.

10. The system according to claim 2, characterized in that, The overall optical path length of the system is ≤10cm, and the system width and height are both no more than 5cm; its ranging range is 0.5m to 200m, the minimum angular resolution is 0.1°, and the ranging accuracy at 200m is ≤2cm.

Images