A dual fast mirror beam pointing stabilization system and method based on matrix decoupling

Abstract

The application relates to the technical field of laser beam stabilization control, and discloses a double fast mirror beam pointing stabilization system and method based on matrix decoupling, wherein a main control module sequentially scans four piezoelectric drive channels to obtain a driving voltage variation, receives a first light spot position, generates a coupling matrix based on the driving voltage variation and the first light spot position, and calculates a decoupling matrix; a light spot detection module collects a second light spot position; the main control module constructs an error vector based on the second light spot position and calculates a control correction amount, maps the control correction amount into a four-way piezoelectric driving voltage increment by using the decoupling matrix, and obtains a target four-way piezoelectric driving voltage; a driving output module performs output processing on the target four-way piezoelectric driving voltage, and outputs the target four-way piezoelectric driving voltage to a double fast mirror execution module to drive the double fast mirror to deflect; and the main control module controls the light spot detection module to perform a next closed-loop control cycle. Through the calibration-decoupling-closed-loop control architecture, the application reduces control crosstalk caused by multi-channel coupling of the double fast mirror, and improves the beam pointing stabilization precision.

Description

Technical Field

[0001] This invention relates to the field of laser beam stabilization and control technology, specifically to a dual fast mirror beam pointing stabilization system and method based on matrix decoupling. Background Technology

[0002] In femtosecond lasers, ultrafast laser amplification, free-space optical communication, high-power laser devices, and precision optical measurement systems, laser beams are affected by factors such as laser thermal drift, cooling system thermal disturbances, folding mirror drift, air turbulence, thermal effects of optical components, and optical platform vibration during transmission, resulting in positional deviations, angular deviations, or position-angle coupling deviations. For long-distance, multi-stage folding optical path systems, even minute angular disturbances can be amplified with propagation distance, causing drift in the final beam spot position and affecting subsequent coupling efficiency, focusing quality, and energy stability. Therefore, ensuring beam pointing stability is a critical issue for ensuring the long-term reliable operation of the system.

[0003] Existing beam pointing stabilization systems typically employ cameras, position-sensitive detectors, or four-quadrant detectors to acquire the beam spot position, and then use piezoelectric ceramic actuators or piezoelectric screws to drive the deflection of fast-reflecting mirrors to achieve beam pointing correction. The dual-fast-reflecting mirror structure is a commonly used architecture, which achieves synchronous adjustment of the beam's spatial position and propagation direction through the coordinated deflection of the first and second fast-reflecting mirrors.

[0004] However, the dual fast-reflecting mirrors are controlled by four separate piezoelectric ceramic drive voltages, and the system outputs four beam positions: near-field X / Y and far-field X / Y. A change in any one voltage, while causing deflection of the corresponding lens, often leads to simultaneous changes in the positions of multiple beams. For example, adjusting the voltage of the X-axis piezoelectric drive channel of the first fast-reflecting mirror will cause significant changes in both the near-field and far-field X-axis due to coupling characteristics. This coupling characteristic, where one change affects multiple others, makes it easy for adjusting one coordinate to cause crosstalk to other coordinates, making it difficult for closed-loop control to quickly and accurately stabilize the beam direction. Summary of the Invention

[0005] This invention provides a dual fast mirror beam pointing stabilization system and method based on matrix decoupling, in order to solve one of the technical problems existing in the prior art.

[0006] In a first aspect, the present invention provides a dual fast mirror beam pointing stabilization system based on matrix decoupling. The system includes a dual fast mirror execution module, a spot detection module, a main control module, and a drive output module. The dual fast mirror execution module has four piezoelectric drive channels. The main control module is used to sequentially scan the four piezoelectric drive channels to obtain the change in drive voltage, receive the first spot position of the near-field and far-field detection surfaces collected by the spot detection module after each scan, generate a coupling matrix based on the change in drive voltage and the first spot position, and calculate the decoupling matrix of the coupling matrix. The spot detection module is also used to collect the positions of the second spot on the near-field and far-field detection surfaces in real time and send them to the main control module; The main control module is also used to construct an error vector based on the position of the second spot, calculate the control correction amount of the error vector, and use a decoupling matrix to map the control correction amount into the four-way piezoelectric drive voltage increment to obtain the target four-way piezoelectric drive voltage. The drive output module is used to process the output of the target four-channel piezoelectric drive voltage and output it to the dual fast mirror execution module; The dual fast-reflecting mirror execution module is used to drive the first and second fast-reflecting mirrors to deflect in coordination based on the target four-channel piezoelectric drive voltage after output processing. The main control module is also used to control the spot detection module to perform the next closed-loop control cycle and re-acquire the second spot positions of the near-field detection surface and the far-field detection surface.

[0007] This invention obtains the coupling matrix through single-channel scanning calibration by the main control module, quantifies the linear coupling relationship between the four piezoelectric driving voltages and the beam position, and calculates the decoupling matrix, providing an accurate mathematical model basis for closed-loop control. In closed-loop control, the beam spot detection module acquires the near-field and far-field beam positions in real time. After constructing the error vector, the main control module calculates its control correction amount, and then uses the decoupling matrix to map the control correction amount to the four driving voltage increments, thereby obtaining the target four piezoelectric driving voltages. After output processing by the drive output module, the output is sent to the dual fast mirror execution module, so that the module drives the first and second fast mirrors to deflect collaboratively based on the output of the target four piezoelectric driving voltages, achieving synchronous correction of beam position and propagation direction. Through the calibration-decoupling-closed-loop control architecture, the control crosstalk caused by multi-channel coupling of the dual fast mirrors can be reduced, improving beam pointing stability accuracy, closed-loop response speed, and embedded real-time control capabilities. It is suitable for femtosecond lasers, ultrafast laser amplification, free-space optical communication, and precision optical measurement systems.

[0008] In one optional implementation, the system further includes a host computer communication module, which is used to receive the target spot coordinates, matrix parameters, control parameters and scanning parameters sent by the host computer during initialization, and send them to the main control module.

[0009] This invention uses a host computer communication module as a parameter configuration channel, which can complete system adaptation without modifying the underlying control program, thus improving the system's versatility and ease of use.

[0010] In one alternative implementation, the main control module includes programmable logic and a processing system; The processing system is used to sequentially select each of the four piezoelectric drive channels as the current scanning channel, and control the drive output module to apply positive and negative stepped drive voltages to the current scanning channel based on the scanning parameters, so that the dual fast mirror execution module deflects the first fast mirror and the second fast mirror according to each drive voltage. The spot detection module is specifically used to acquire the first spot position of the near-field detection surface and the far-field detection surface after each driving voltage is applied, and send it to the main control module; Programmable logic is used to calculate the position of the first spot for each driving voltage to obtain multiple first spot coordinates, including near-field X-axis spot coordinates, near-field Y-axis spot coordinates, far-field X-axis spot coordinates and far-field Y-axis spot coordinates; The processing system is also used to calculate the changes in multiple spot coordinates based on the reference coordinates and the coordinates of multiple first spot for each driving voltage, and to calculate the changes in driving voltage based on the reference voltage and each driving voltage in the scanning parameters. The processing system is also used to perform linear fitting on the changes in the coordinates of each spot based on all sampling points, obtain multiple fitting slopes, combine them into the coupling coefficient of the current scanning channel, and combine all the coupling coefficients of the four piezoelectric drive channels into a coupling matrix. Each sampling point includes the change in the driving voltage corresponding to a driving voltage and multiple changes in the coordinates of the spot.

[0011] This invention utilizes a single-channel bidirectional stepped scanning processing system to separate the independent influence of each channel on the beam spot position, facilitating the accurate establishment of the correspondence between input and output, and eliminating the impact of hysteresis on calibration accuracy. Furthermore, by linearly fitting the beam spot coordinate changes and driving voltage changes, a coupling coefficient is obtained, leading to a coupling matrix. This matrix quantifies the influence of the voltages of the four piezoelectric driving channels on the four beam spot coordinates, thereby improving the accuracy of coordinated correction of beam position and angle deviations during subsequent closed-loop control without relying on ideal geometric models or fixed empirical parameters.

[0012] In an alternative implementation, before performing a linear fit on the coordinate changes of each spot based on all sampling points, the processing system is further configured to: Based on all sampling points, linear fitting was performed on the coordinate changes of each spot to obtain multiple fitted straight lines; For each sampling point, calculate the residual between the sampling point and each fitted line; Calculate the mean and standard deviation based on the residuals of all sampling points; Based on the mean and standard deviation, determine whether each sampling point is abnormal; Remove sampling points that are judged to be abnormal.

[0013] This invention identifies and removes abnormal sampling points by residuals, enabling the processing system to re-fit linearly using the remaining valid sampling points and ultimately generate a coupling matrix. This reduces the impact of local abnormal data on the coupling coefficient, thereby improving the identification accuracy and reliability of the coupling matrix.

[0014] In one alternative implementation, the processing system is further configured to perform a regularized pseudo-inverse on the coupling matrix to obtain a decoupling matrix.

[0015] This invention obtains the decoupling matrix by performing a regularized pseudo-inverse on the coupling matrix, which achieves a trade-off between the accuracy of the ideal solution and the smoothness of the control quantity, thereby improving the numerical stability and engineering robustness of decoupled control.

[0016] In one optional implementation, the programmable logic is further used to calculate the position of the second spot to obtain multiple second spot coordinates; The processing system is also used to calculate the error vector based on the coordinates of multiple second spot and the target spot, and to use a closed-loop control algorithm to calculate the control correction amount of the error vector based on the matrix parameters. The control correction amount is then multiplied with the decoupling matrix to obtain the four-channel piezoelectric drive voltage increment. The processing system is also used to superimpose the incremental values ​​of the four piezoelectric drive voltages onto the four piezoelectric drive voltages output by the drive output module in the previous closed-loop control cycle to obtain the target four piezoelectric drive voltages.

[0017] This invention uses programmable logic to calculate the beam coordinates in real time. The processing system calculates the control correction based on the error vector and maps it to the voltage increment, which is then superimposed on the output value of the previous cycle to ensure continuous and smooth voltage adjustment, thus achieving rapid, accurate, and stable correction of the beam direction.

[0018] In one alternative implementation, the processing system is further configured to continuously determine whether the decoupling matrix has failed, and regenerate the coupling matrix when the decoupling matrix fails.

[0019] This invention continuously monitors the closed-loop control performance, enabling timely detection of decoupling matrix failures and triggering recalibration to update the coupling and decoupling matrices. This ensures that the decoupling matrix always matches the current system state, avoiding the risk of beam divergence caused by continuing to operate under a failed matrix, and improving the long-term reliability and adaptability of the system.

[0020] In one optional implementation, the drive output module includes an adjustment unit, a digital-to-analog conversion unit, and an amplification unit; The adjustment unit is used to smooth the target four-channel piezoelectric drive voltage; The digital-to-analog converter unit is used to convert the smoothed target four piezoelectric drive voltages into four analog voltage signals; The amplification unit is used to amplify the four analog voltage signals and output the amplified four analog voltage signals to the dual fast-reflection mirror execution module.

[0021] This invention performs smoothing processing through an adjustment unit, then converts the digital control quantity into an analog voltage signal, and finally amplifies it to achieve a complete conversion link from the digital control quantity of the main control module to the actual drive signal of the actuator, ensuring the accurate execution of control commands.

[0022] In one alternative implementation, the processing system is also configured to continuously monitor the coordinates of multiple second light spots and the four analog voltage signals before and after amplification based on control parameters, and to perform corresponding protection operations when an anomaly is detected.

[0023] This invention improves the safety and reliability of the system's long-term operation by continuously monitoring for abnormal conditions and executing protective actions.

[0024] Secondly, this invention provides a method for stabilizing the beam pointing of a dual fast-reflecting mirror based on matrix decoupling, the method comprising: The driving voltage change is obtained by sequentially scanning the four piezoelectric driving channels. The first spot position of the near-field and far-field detection surfaces is received after each scan. A coupling matrix is ​​generated based on the driving voltage change and the first spot position. The decoupling matrix of the coupling matrix is ​​calculated. Real-time acquisition of the second spot positions of the near-field and far-field detection surfaces; An error vector is constructed based on the position of the second spot. The control correction amount of the error vector is calculated, and the control correction amount is mapped to the four piezoelectric drive voltage increments using a decoupling matrix to obtain the target four piezoelectric drive voltages. The output voltages of the four target piezoelectric drive channels are processed. Based on the target four-channel piezoelectric driving voltage after output processing, the first fast mirror and the second fast mirror are driven to deflect in coordination. The next closed-loop control cycle begins, returning to the step of real-time acquisition of the second spot positions of the near-field and far-field detection surfaces.

[0025] This invention obtains the coupling matrix through single-channel scanning calibration, quantifies the linear coupling relationship between the four piezoelectric driving voltages and the beam spot position, and calculates the decoupling matrix, providing an accurate mathematical model basis for closed-loop control. In closed-loop control, the near-field and far-field beam spot positions are acquired in real time, and after constructing the error vector, the control correction is calculated. Then, the decoupling matrix is ​​used to map the control correction to the four driving voltage increments, thereby obtaining the target four piezoelectric driving voltages. After output processing, the first and second fast-reflecting mirrors are driven to deflect collaboratively, achieving synchronous correction of the beam position and propagation direction. Through the calibration-decoupling-closed-loop control architecture, control crosstalk caused by multi-channel coupling of dual fast-reflecting mirrors can be reduced, improving beam pointing stability accuracy, closed-loop response speed, and embedded real-time control capabilities. Attached Figure Description

[0026] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0027] Figure 1 This is a schematic diagram of a dual fast-reflecting mirror beam pointing stabilization system based on matrix decoupling according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the decoupling process of the closed-loop control process according to an embodiment of the present invention; Figure 3 This is a schematic flowchart of a dual fast-reflecting mirror beam pointing stabilization system based on matrix decoupling according to an embodiment of the present invention; Figure 4 This is a flowchart of a dual fast mirror beam pointing stabilization method based on matrix decoupling according to an embodiment of the present invention. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0030] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0031] Figure 1 This is a schematic diagram of a dual-fast-reflecting mirror beam pointing stabilization system based on matrix decoupling according to an embodiment of the present invention, as shown below. Figure 1 As shown, the system includes a dual fast-reflection mirror execution module, a spot detection module, a main control module, and a drive output module. The dual fast-reflection mirror execution module has four piezoelectric drive channels. The main control module is used to sequentially scan the four piezoelectric drive channels to obtain the drive voltage change, receive the first spot positions of the near-field and far-field detection surfaces acquired by the spot detection module after each scan, and generate a coupling matrix based on the drive voltage change and the first spot positions, and calculate the decoupling matrix of the coupling matrix. The spot detection module is also used to acquire the second spot positions of the near-field and far-field detection surfaces in real time and send them to the main control module. The main control module also uses... An error vector is constructed based on the position of the second light spot. The control correction amount of the error vector is calculated, and a decoupling matrix is ​​used to map the control correction amount into the increment of the four piezoelectric driving voltages to obtain the target four piezoelectric driving voltages. The drive output module is used to process the target four piezoelectric driving voltages and output them to the dual fast mirror execution module. The dual fast mirror execution module is used to drive the first fast mirror and the second fast mirror to deflect in coordination based on the processed target four piezoelectric driving voltages. The main control module is also used to control the light spot detection module to perform the next closed-loop control cycle and re-acquire the positions of the second light spots on the near-field detection surface and the far-field detection surface.

[0032] Specifically, the dual fast-reflecting mirror system includes a first fast-reflecting mirror and a second fast-reflecting mirror. These two mirrors are arranged sequentially along the optical path, forming a folded dual fast-reflecting mirror optical path. The light beam is reflected sequentially by the first and second fast-reflecting mirrors before exiting. Each fast-reflecting mirror is driven by two orthogonal piezoelectric ceramic actuators, namely an X-axis piezoelectric ceramic actuator and a Y-axis piezoelectric ceramic actuator, which control the deflection angle of the mirror surface along the X and Y axes, respectively. Therefore, the dual fast-reflecting mirror actuator module has four independent piezoelectric drive channels: the X-axis and Y-axis piezoelectric drive channels for the first fast-reflecting mirror, and the X-axis and Y-axis piezoelectric drive channels for the second fast-reflecting mirror. Through the combined adjustment of the four piezoelectric drive voltages, the deflection of the first fast-reflecting mirror mainly changes the lateral position of the light beam on the near-field detection surface, while the deflection of the second fast-reflecting mirror mainly changes the propagation angle of the light beam on the far-field detection surface. The coordinated deflection of both mirrors can simultaneously achieve synchronous correction of the spatial position and propagation direction of the light beam, avoiding the problem of not being able to distinguish between positional and angular offsets when relying on only a single detection surface.

[0033] During the initial calibration phase, the main control module sequentially scans each of the four piezoelectric drive channels to obtain the changes in drive voltage. Scanning refers to applying drive voltage to the piezoelectric drive channels, causing the dual fast-reflecting mirrors to deflect and alter the spot position. Therefore, after each scan by the main control module, the spot detection module acquires the first spot position on both the near-field and far-field detection surfaces. This allows the main control module to generate a coupling matrix based on the change in drive voltage and the first spot position, quantifying the linear coupling relationship between the four piezoelectric drive voltages and the spot position. Furthermore, based on the coupling matrix, a decoupling matrix is ​​calculated, enabling the spot error to be mapped back to the drive voltage increment.

[0034] During the closed-loop control phase, the spot detection module acquires the second spot positions on the near-field and far-field detection surfaces in real time and sends them to the main control module. The main control module constructs an error vector based on the real-time second spot positions and calculates the control correction amount, representing the direction and magnitude of movement required to return the spot coordinates to the target position. A decoupling matrix is ​​used to map the control correction amount to the increment of each piezoelectric drive voltage, which is then superimposed on the four piezoelectric drive voltages output by the drive output module in the previous closed-loop control cycle to obtain the current target four piezoelectric drive voltages. After output processing by the drive output module, the output is sent to the dual fast-reflecting mirror execution module. This module, based on the processed target four piezoelectric drive voltages, drives the first and second fast-reflecting mirrors to deflect collaboratively, changing the beam propagation direction. This corrects the lateral position deviation of the spot on the near-field detection surface and the propagation angle deviation on the far-field detection surface, suppressing crosstalk between multiple channels and solving the problem of slowed closed-loop response and increased steady-state error caused by one-to-many movement. The main control module controls the spot detection module to reacquire the spot positions of the near-field and far-field detection surfaces, and continues closed-loop control to ensure the stability of the beam pointing.

[0035] In some optional implementations, the system also includes a host computer communication module, which receives the target spot coordinates, matrix parameters, control parameters and scanning parameters sent by the host computer during initialization, and sends them to the main control module.

[0036] Specifically, the host computer communication module is used to realize parameter interaction between the system and the user. The user can configure the target spot coordinates, matrix parameters, control parameters, and scanning parameters through the host computer. Among them, the target spot coordinates represent the near-field and far-field target positions that the desired spot should reach during the closed-loop control stage; the scanning parameters represent the voltage range, step value, and waiting time per step for single-channel scanning during the calibration stage, which are used to control the scanning execution process; the matrix parameters represent the proportional, integral, and derivative gain matrices in the closed-loop control algorithm, which are used to calculate the control correction; the control parameters include the anomaly monitoring threshold.

[0037] The host computer communication module sends the configured parameters to the main control module, enabling the main control module to perform calibration and closed-loop control according to the user-defined goals, rules, and safety boundaries. Furthermore, the host computer communication module allows for flexible parameter configuration based on different application scenarios and accuracy requirements without modifying the underlying control program, enabling rapid system adaptation to different optical path structures and control needs.

[0038] In some optional implementations, the main control module includes programmable logic and a processing system; the processing system is used to sequentially select each of the four piezoelectric drive channels as the current scanning channel, and control the drive output module to apply forward and reverse stepped drive voltages to the current scanning channel based on the scanning parameters, so that the dual fast mirror execution module deflects the first and second fast mirrors according to each drive voltage; the spot detection module is specifically used to collect the first spot positions of the near-field detection surface and the far-field detection surface after each drive voltage is applied, and send them to the main control module; the programmable logic is used to calculate the first spot position of each drive voltage to obtain multiple first spot coordinates, and multiple first spots The coordinates include near-field X-axis spot coordinates, near-field Y-axis spot coordinates, far-field X-axis spot coordinates, and far-field Y-axis spot coordinates. The processing system is also used to calculate multiple spot coordinate changes based on the reference coordinates and multiple first spot coordinates for each driving voltage, and to calculate the driving voltage change based on the reference voltage and each driving voltage in the scanning parameters. The processing system is also used to perform linear fitting on each spot coordinate change based on all sampling points to obtain multiple fitting slopes, which are combined into the coupling coefficient of the current scanning channel. All coupling coefficients of the four piezoelectric driving channels are combined into a coupling matrix. Each sampling point includes the driving voltage change corresponding to a driving voltage and multiple spot coordinate changes.

[0039] Specifically, the main control module can be a ZYNQ chip, which includes a processing system (PS) and programmable logic (PL). During the calibration phase, the processing system sequentially selects each of the four piezoelectric drive channels as the current scanning channel. According to the scanning parameters (e.g., voltage range 0~5V, step value 0.5V, wait time per step 50ms), it applies forward and reverse stepped drive voltages to the current scanning channel, while the other channels maintain a constant reference voltage. This single-channel scanning method isolates the independent influence of a single channel on the spot position, facilitating the accurate establishment of the correspondence between input and output. Simultaneously, the use of bidirectional forward and reverse scanning allows for the full acquisition of response data from the piezoelectric ceramic during voltage boosting and bucking, eliminating the impact of hysteresis on calibration accuracy.

[0040] Taking the X-axis piezoelectric drive channel of the first fast-reflecting mirror as the current scanning channel as an example: The processing system first applies a 0V drive voltage to the X-axis piezoelectric drive channel after processing by the drive output module. After a 50ms waiting period to reduce the impact of transient oscillations of the piezoelectric actuator, asynchronous camera exposure, or unstable light spot, the light spot detection module acquires the current position of the first light spot. Subsequently, the processing system increases the drive voltage to 0.5V, processes it again through the drive output module, and applies it to the X-axis piezoelectric drive channel, causing a corresponding deflection of the X-axis of the first fast-reflecting mirror, and displacement of the light spot on the detection surface. After a 50ms waiting period to stabilize, the position of the first light spot is acquired again. This process of gradually increasing the voltage is repeated, recording the position of the first light spot at each drive voltage of 0V, 0.5V, 1.0V, 1.5V...5.0V, completing the forward scan. The processing system then gradually reduces the voltage starting from 5.0V, following the sequence of 5.0V, 4.5V, 4.0V...0V. At each driving voltage, it waits 50ms and acquires the position of the first light spot to complete the reverse scan.

[0041] After applying each driving voltage, the spot detection module acquires the first spot position on the near-field and far-field detection surfaces and sends it to the main control module. If the spot detection module is a two-channel position-sensitive detector connected via an HD15 interface, the original signal range can be ±10V. The output first spot position includes the near-field and far-field X-axis differential signals, Y-axis differential signals, and the sum of light intensity signals. The programmable logic performs normalized differential processing on this original signal, that is, the ratio of the X-axis differential signal to the sum of light intensity signals is used as the X-axis position quantity, and the ratio of the Y-axis differential signal to the sum of light intensity signals is used as the Y-axis position quantity, ultimately obtaining multiple first spot coordinates, including near-field X-axis spot coordinates, near-field Y-axis spot coordinates, far-field X-axis spot coordinates, and far-field Y-axis spot coordinates. This calculation method can effectively reduce the impact of light intensity fluctuations on coordinate calculation, making the output coordinates mainly reflect changes in spot position rather than changes in light power. If the spot detection module is a 2-channel CameraLink industrial camera, the programmable logic calculates the coordinates of the multiple first spots using an image grayscale weighted centroid algorithm.

[0042] The reference coordinates are the spot coordinates acquired by the four piezoelectric drive channels under the reference voltage during initialization. Based on the reference coordinates and multiple first spot coordinates for each drive voltage, the processing system calculates the changes in multiple spot coordinates corresponding to that drive voltage, including coordinate changes in four dimensions: near-field X-axis, near-field Y-axis, far-field X-axis, and far-field Y-axis. Simultaneously, the processor calculates the drive voltage change between the reference voltage and the drive voltage in the scanning parameters. For example, if the current drive voltage is 2.5V and the reference voltage is 0V, the drive voltage change is +2.5V; if the near-field X-axis spot coordinate under the current drive voltage is 105μm and the reference coordinate is 100μm, the near-field X-axis spot coordinate change is +5μm.

[0043] The change in driving voltage corresponding to a single driving voltage and the changes in multiple spot coordinates are used as a sampling point. After all sampling points are collected, the processor uses the change in driving voltage as the independent variable and performs linear fitting on the changes in spot coordinates for each, obtaining four fitted lines, thus yielding four fitting slopes. For example, fitting the change in driving voltage corresponding to each driving voltage of the current scanning channel with all near-field X-axis spot coordinate changes corresponding to each driving voltage yields a fitting slope, which represents the influence of the voltage change of the current scanning channel on the near-field X-axis spot. Similarly, the fitting slopes for the near-field Y, far-field X, and far-field Y are obtained. These four fitting slopes are combined to form the coupling coefficient of the current scanning channel, serving as a column of the coupling matrix. After all four piezoelectric driving channels are scanned, the coupling coefficients of all channels are combined to form a complete 4×4 coupling matrix, representing the influence relationship between the voltage of the four piezoelectric driving channels and the four spot coordinates.

[0044] In a dual-fast-reflecting mirror system, the four piezoelectric drive channels are cross-coupled with the four beam coordinates. This means that a voltage change in any piezoelectric drive channel not only affects its primary beam coordinate but may also cause changes in other near-field or far-field beam coordinates. If each axis error is simply mapped to a single drive channel for independent correction, adjusting one channel can easily generate non-major-axis crosstalk to other channels, leading to slower closed-loop response and increased steady-state error. By obtaining the coupling matrix through calibration, the influence of each drive channel on each beam coordinate under actual assembly and adjustment conditions can be quantified. This allows for the simultaneous calculation of the optimal distribution of the four piezoelectric drive voltages during subsequent closed-loop control, without relying on ideal geometric models or fixed empirical parameters, thus improving the accuracy of coordinated correction of beam position and angle deviations.

[0045] In some optional implementations, before performing linear fitting on the coordinate changes of each spot based on all sampling points, the processing system is further configured to: perform linear fitting on the coordinate changes of each spot based on all sampling points to obtain multiple fitted lines; for each sampling point, calculate the residual between the sampling point and each fitted line; calculate the mean and standard deviation based on the residuals of all sampling points; determine whether each sampling point is abnormal based on the mean and standard deviation; and remove sampling points determined to be abnormal.

[0046] Specifically, before generating the coupling matrix, the processing system first uses the change in driving voltage as the independent variable to perform linear fitting on the changes in the four spot coordinates in the near-field X-axis, near-field Y-axis, far-field X-axis, and far-field Y-axis, respectively, obtaining four fitted lines. For each sampling point, the change in driving voltage at that sampling point is substituted into the four fitted lines to obtain the predicted value of the change in spot coordinates in each dimension. Then, the difference between the change in spot coordinates at that sampling point and the predicted value is calculated, which is the residual of that sampling point in that dimension. The larger the residual, the higher the probability that the sampling point is affected by noise or disturbance. Thus, the residuals of each sampling point in the four dimensions of near-field X-axis, near-field Y-axis, far-field X-axis, and far-field Y-axis can be obtained.

[0047] For the same dimension, calculate the mean and standard deviation of all residuals in that dimension. For each sampling point, if the following conditions are met... If the sampled point is an outlier, it will be removed. This represents the residual of the sampling point in the i-th dimension; This represents the mean of the i-th dimension; Let represent the standard deviation of the i-th dimension; n can be 2 or 3, preferably 3. It should be noted that each sampling point includes residuals in four dimensions, and if the residual of any dimension satisfies the above formula, then the sampling point is regarded as an outlier and removed.

[0048] By identifying and eliminating abnormal sampling points introduced by noise or disturbances, such as instantaneous camera noise, spot recognition errors, mechanical disturbances, sudden changes in light intensity, or communication sampling, the processing system can re-fit linearly using the remaining valid sampling points to ultimately generate a coupling matrix. This reduces the impact of local abnormal data on the coupling coefficient, thereby improving the identification accuracy and reliability of the coupling matrix.

[0049] In some alternative implementations, the processing system is also used to perform a regularized pseudo-inverse on the coupling matrix to obtain a decoupling matrix.

[0050] Specifically, to avoid amplifying measurement noise by directly inverting the matrix, a Tikhonov regularized pseudo-inverse is used to solve for the decoupling matrix of the coupling matrix:

[0051] In the formula, Represents the decoupling matrix; denoted by ; I represents the identity matrix; B represents the coupling matrix.

[0052] By introducing a regularization parameter, the amplification of control input caused by ill-conditioned matrices, calibration errors, and detection noise can be suppressed during matrix inversion. When the coupling matrix has a large condition number, similar channel responses, or perturbations in the calibration data, direct inversion can convert even small spot measurement errors into large driving voltage changes, leading to over-action or closed-loop oscillation. By using a regularized pseudo-inverse, a trade-off can be achieved between the accuracy of the ideal solution and the smoothness of the control input, improving the numerical stability and engineering robustness of decoupled control.

[0053] In some alternative implementations, such as Figure 2 As shown, the programmable logic is also used to solve the position of the second spot to obtain multiple second spot coordinates; the processing system is also used to calculate the error vector based on the multiple second spot coordinates and the target spot coordinates, and to use a closed-loop control algorithm to calculate the control correction amount of the error vector based on the matrix parameters, and to perform matrix multiplication operation on the control correction amount and the decoupling matrix to obtain the four-channel piezoelectric drive voltage increment; the processing system is also used to superimpose the four-channel piezoelectric drive voltage increments onto the four-channel piezoelectric drive voltage output by the drive output module in the previous closed-loop control cycle to obtain the target four-channel piezoelectric drive voltage.

[0054] Specifically, the programmable logic calculates the position of the second light spot using the same calculation method as the first light spot position, obtaining multiple second light spot coordinates, including near-field X-axis light spot coordinates, near-field Y-axis light spot coordinates, far-field X-axis light spot coordinates, and far-field Y-axis light spot coordinates, which are used to represent the actual position and propagation angle of the beam at the current moment. The processing system calculates the difference between the target light spot coordinates and the second light spot coordinates, constructing a four-dimensional error vector:

[0055] In the formula, This represents the error vector at time k; This represents the error in the near-field X direction; This represents the error in the near-field Y direction; This represents the error in the far-field X direction; This represents the error in the far-field Y direction.

[0056] Among them, the errors in the near-field X and near-field Y directions represent the deviation of the beam position, while the errors in the far-field X and far-field Y directions represent the deviation of the beam propagation angle.

[0057] The closed-loop control algorithm can be either an incremental matrix PI algorithm or an incremental matrix PID control algorithm. If an incremental matrix PI algorithm is used, the control correction amount for the error vector can be calculated using the following formula:

[0058] In the formula, This represents the control correction amount for the k-loop control cycle; This represents the proportional gain matrix, which is determined by matrix parameters. This represents the error vector of the k-1 closed-loop control cycle, which is also the error vector calculated in the previous closed-loop control cycle. This represents the integral gain matrix, which is determined by the matrix parameters.

[0059] Among them, the proportional term Reflects the trend of error change and is used for rapid response to sudden shifts in the light spot; integral term It reflects the cumulative effect of error and is used to eliminate steady-state residual error, ensuring that the light spot converges accurately to the target light spot coordinates.

[0060] If an incremental matrix PID control algorithm is used, a derivative term is added to the above formula to further suppress overshoot and oscillation. The closed-loop control algorithm described above can correct the output voltage cycle by cycle based on the current error, the error calculated in the previous closed-loop control cycle, and the error integral term, instead of directly re-applying the absolute drive voltage each cycle. This reduces the risks caused by sudden changes in control output and integral accumulation.

[0061] The processing system performs matrix multiplication on the control correction quantity and the decoupling matrix to obtain the four piezoelectric drive voltage increments, which are the drive voltage changes corresponding to the four piezoelectric drive channels, respectively corresponding to the X-axis, Y-axis, X-axis, and Y-axis of the first fast reflector, the second fast reflector, and the second fast reflector. This retains the ability of PI / PID control to adjust steady-state error and dynamic response, while also suppressing crosstalk between the four execution channels through matrix decoupling, achieving multi-channel decoupled control and solving the problem of one-to-many control.

[0062] The processing system superimposes the incremental values ​​of the four piezoelectric drive voltages onto the four piezoelectric drive voltages output by the drive output module in the previous closed-loop control cycle to obtain the target four piezoelectric drive voltages. For example, if the voltage output of the X-axis piezoelectric drive channel of the first fast-reflecting mirror in the previous closed-loop control cycle is 20V, and the current calculated voltage increment of this channel is +0.5V, then the voltage of this channel after superposition is 20.5V.

[0063] In some alternative implementations, the processing system is also used to continuously determine whether the decoupling matrix has failed, and to regenerate the coupling matrix when the decoupling matrix fails.

[0064] Specifically, during closed-loop operation, the processing system continuously monitors the effectiveness of the decoupling matrix. When it is determined that the decoupling matrix can no longer accurately reflect the coupling relationship of the current system, a recalibration process is triggered to update the coupling and decoupling matrices. The judgment conditions for the decoupling matrix failure include at least one of the following: Condition 1, the near-field / far-field error has not decreased within N consecutive closed-loop control cycles, or the root mean square value of the near-field / far-field error is greater than a preset threshold; Condition 2, the target four-channel piezoelectric drive voltage has reached the limit value but the error vector continues to increase; Condition 3, the actual spot change direction is inconsistent with the direction predicted by the decoupling matrix, and the prediction error exceeds a preset threshold; Condition 4, the condition number of the coupling matrix exceeds a preset threshold, indicating that the matrix is ​​highly ill-conditioned.

[0065] When the processing system determines that the current decoupling matrix is ​​invalid, it pauses the closed-loop control and re-executes the calibration to generate a new coupling matrix and decoupling matrix.

[0066] Optionally, in addition to the failure detection of the decoupling matrix mentioned above, the processing system can also regenerate the coupling and decoupling matrices when optical path adjustment, spot cross-boundary, or system power-on is detected, to ensure that the calibration parameters always match the current system state. The criteria for determining optical path adjustment include at least one of the following: the user actively triggers recalibration by reissuing a calibration command through the host computer; the target spot coordinates change; the reference coordinates undergo a step change; the near-field and far-field spot coordinates simultaneously show significant shifts, and the shift exceeds a preset threshold. The criteria for determining spot cross-boundary include at least one of the following: the centroid coordinates exceed the effective image area; the distance between the spot centroid and the image boundary is less than a preset safety margin; the light intensity and signal of the position-sensitive detector are lower than the effective light intensity threshold; the coordinate calculation result exceeds the calibration range.

[0067] In some optional implementations, the drive output module includes an adjustment unit, a digital-to-analog converter, and an amplification unit; the adjustment unit is used to smooth the target four piezoelectric drive voltages; the digital-to-analog converter is used to convert the smoothed target four piezoelectric drive voltages into four analog voltage signals; the amplification unit is used to amplify the four analog voltage signals and output the amplified four analog voltage signals to the dual fast-reflection mirror execution module.

[0068] Specifically, in the closed-loop control process, after the main control module calculates the target four piezoelectric drive voltages, it sends them to the adjustment unit for smoothing, including digital filtering or slope limiting. This suppresses sudden changes in control quantities, avoids shocks caused by drastic jumps in output voltage, reduces the risk of mechanical oscillation, and extends the service life of the piezoelectric ceramic actuators in the dual fast-reflecting mirror actuator module. After smoothing, the signals are sent to the digital-to-analog converter unit, which is a four-channel digital-to-analog converter. This unit converts the digital form of the four piezoelectric drive voltages into four analog voltage signals, typically 0~5V. The amplification unit is a PA83 high-voltage operational amplifier, which amplifies the four low-voltage analog voltage signals, for example, by approximately 20 times gain to 0~100V with a resolution accuracy better than 10mV, generating a high-voltage drive signal sufficient to drive the piezoelectric ceramic actuators. The amplified four analog voltage signals are then output to the dual fast-reflecting mirror actuator module, applied to the corresponding four piezoelectric drive channels, driving the first and second fast-reflecting mirrors to produce coordinated deflection. This changes the beam propagation path, corrects beam pointing deviation, and achieves synchronous correction of near-field position deviation and far-field angle deviation.

[0069] In some optional implementations, the processing system is also used to continuously monitor the coordinates of multiple second light spots and the four analog voltage signals before and after amplification based on control parameters, and to perform corresponding protection operations when an anomaly is detected.

[0070] Specifically, during closed-loop control, the processing system continuously monitors control parameters, including the coordinates of multiple second light spots and the four analog voltage signals before and after amplification. The monitoring includes: whether the four analog voltage signals before and after amplification exceed preset limits; whether the centroid coordinates of the light spot exceed the effective detection area of ​​the camera or position-sensitive detector; and whether the light spot intensity is below the effective recognition threshold. When any of the above abnormalities is detected, the processing system executes corresponding protection operations based on the type of abnormality. These protection operations include at least one of the following: output hold (maintaining the four amplified analog voltage signals unchanged); output limiting (forcibly limiting the four analog voltage signals within a safe range); high-voltage drive shutdown (cutting off the output of the amplification unit to avoid damaging the piezoelectric ceramic actuator); closed-loop suspension (stopping closed-loop control and resuming it after the abnormality is resolved); alarm upload (sending fault information to the host computer via the host computer communication module to notify user intervention); and recalibration (triggering the process of regenerating the coupling and decoupling matrices).

[0071] In some alternative implementations, Figure 3 This is a schematic flowchart of a dual-fast-reflecting mirror beam pointing stabilization system based on matrix decoupling according to an embodiment of the present invention, as shown below. Figure 3As shown, the system is powered on and initialized, entering the calibration phase. First, a single-channel bidirectional stepped scan is used to acquire the position of the first spot after each scan, calculating its coordinates. The changes in spot coordinates and driving voltage are calculated, and after anomaly removal, linear fitting is performed to generate a coupling matrix. This matrix is ​​then regularized and pseudo-inversed to obtain the decoupling matrix. In the closed-loop control phase, the position of the second spot is acquired in real-time, its coordinates are calculated, and the error vector between the second spot and the target spot coordinates is calculated. The control correction is calculated using the closed-loop control algorithm and mapped to the four-channel piezoelectric drive voltage increment using the decoupling matrix, thus obtaining the target four-channel piezoelectric drive voltage. The target four-channel piezoelectric drive voltage is adjusted, converted from digital to analog, and amplified to drive the dual fast-reflecting mirrors through the amplified four-channel analog voltage signals, completing the current closed-loop control cycle. Simultaneously, protection operations are executed when an anomaly is detected. The system determines whether to continue closed-loop control. If it continues, it returns to the step of acquiring the second spot position in real-time; if the coupling matrix fails, recalibration is performed; if closed-loop control is stopped, the system stops operating.

[0072] This invention obtains the coupling matrix through single-channel scanning calibration by the main control module, quantifies the linear coupling relationship between the four piezoelectric driving voltages and the beam position, and calculates the decoupling matrix, providing an accurate mathematical model basis for closed-loop control. In closed-loop control, the beam spot detection module acquires the near-field and far-field beam positions in real time. After constructing the error vector, the main control module calculates its control correction amount, and then uses the decoupling matrix to map the control correction amount to the four driving voltage increments, thereby obtaining the target four piezoelectric driving voltages. After output processing by the drive output module, the output is sent to the dual fast mirror execution module, so that the module drives the first and second fast mirrors to deflect collaboratively based on the output of the target four piezoelectric driving voltages, achieving synchronous correction of beam position and propagation direction. Through the calibration-decoupling-closed-loop control architecture, the control crosstalk caused by multi-channel coupling of the dual fast mirrors can be reduced, improving beam pointing stability accuracy, closed-loop response speed, and embedded real-time control capabilities. It is suitable for femtosecond lasers, ultrafast laser amplification, free-space optical communication, and precision optical measurement systems.

[0073] According to an embodiment of the present invention, a method for stabilizing beam pointing of dual fast mirrors based on matrix decoupling is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0074] This embodiment provides a matrix decoupling-based method for beam pointing stabilization of dual fast mirrors, which can be used in the aforementioned matrix decoupling-based dual fast mirror beam pointing stabilization system. Figure 4This is a flowchart of a dual-fast-reflecting mirror beam pointing stabilization method based on matrix decoupling according to an embodiment of the present invention, as follows: Figure 4 As shown, the process includes the following steps: Step S401: Scan the four piezoelectric drive channels sequentially to obtain the change in drive voltage, receive the first spot position of the near-field and far-field detection surfaces after each scan, generate a coupling matrix based on the change in drive voltage and the first spot position, and calculate the decoupling matrix of the coupling matrix.

[0075] Specifically, the main control module performs a single-channel stepped scan on each of the four piezoelectric drive channels sequentially, recording the change in drive voltage and the position of the first light spot in each scan. This generates a coupling matrix to quantify the linear coupling relationship between the four drive voltages and the light spot positions. A decoupling matrix is ​​then calculated based on the coupling matrix, serving as the decoupling parameter for closed-loop control.

[0076] Step S402: Real-time acquisition of the second spot positions of the near-field and far-field detection surfaces.

[0077] Specifically, after the closed-loop control begins, the positions of the second light spots on the near-field and far-field detection surfaces are collected in real time to provide a real-time data basis for the closed-loop control.

[0078] Step S403: Construct an error vector based on the position of the second spot, calculate the control correction amount of the error vector, and use a decoupling matrix to map the control correction amount into the four-way piezoelectric drive voltage increment to obtain the target four-way piezoelectric drive voltage.

[0079] Specifically, based on the coordinates of the second spot, an error vector between it and the target position is constructed, the control correction is calculated, and the control correction is mapped to the four-way piezoelectric drive voltage increment using the decoupling matrix, thus obtaining the target four-way piezoelectric drive voltage for the current closed-loop control cycle.

[0080] Step S404: Output processing of the target four-channel piezoelectric drive voltage.

[0081] Specifically, the target four-channel piezoelectric drive voltage is processed to be safely and accurately converted into signals that drive the fast-reflecting mirror deflection.

[0082] Step S405: Based on the target four-channel piezoelectric drive voltage after output processing, drive the first fast mirror and the second fast mirror to deflect in coordination.

[0083] Specifically, based on the four high-voltage drive signals after output processing, the piezoelectric drive channels of the first and second fast mirrors are driven to generate corresponding deflections, thereby changing the propagation direction of the beam and correcting the lateral position deviation of the light spot on the near-field detection surface and the propagation angle deviation on the far-field detection surface.

[0084] Step S406: Proceed to the next closed-loop control cycle and return to the step of real-time acquisition of the second spot positions of the near-field and far-field detection surfaces.

[0085] Specifically, the next closed-loop control cycle is performed, returning to step S402 to form a continuous closed-loop control cycle, ensuring the long-term stability of the beam direction.

[0086] This invention obtains the coupling matrix through single-channel scanning calibration, quantifies the linear coupling relationship between the four piezoelectric driving voltages and the beam spot position, and calculates the decoupling matrix, providing an accurate mathematical model basis for closed-loop control. In closed-loop control, the near-field and far-field beam spot positions are acquired in real time, and after constructing the error vector, the control correction is calculated. Then, the decoupling matrix is ​​used to map the control correction to the four driving voltage increments, thereby obtaining the target four piezoelectric driving voltages. After output processing, the first and second fast-reflecting mirrors are driven to deflect collaboratively, achieving synchronous correction of the beam position and propagation direction. Through the calibration-decoupling-closed-loop control architecture, control crosstalk caused by multi-channel coupling of dual fast-reflecting mirrors can be reduced, improving beam pointing stability accuracy, closed-loop response speed, and embedded real-time control capabilities.

[0087] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A beam pointing stabilization system based on matrix decoupling using dual fast mirrors, characterized in that, The system includes a dual fast-reflection mirror execution module, a spot detection module, a main control module, and a drive output module. The dual fast-reflection mirror execution module has four piezoelectric drive channels. The main control module is used to sequentially scan the four piezoelectric drive channels to obtain the change in drive voltage, receive the first spot position of the near-field detection surface and the far-field detection surface collected by the spot detection module after each scan, generate a coupling matrix based on the change in drive voltage and the first spot position, and calculate the decoupling matrix of the coupling matrix. The spot detection module is also used to collect the second spot positions of the near-field detection surface and the far-field detection surface in real time and send them to the main control module; The main control module is also used to construct an error vector based on the position of the second spot, calculate the control correction amount of the error vector, and use the decoupling matrix to map the control correction amount into the four-channel piezoelectric drive voltage increment to obtain the target four-channel piezoelectric drive voltage. The drive output module is used to process the target four-channel piezoelectric drive voltage and output it to the dual fast mirror execution module. The dual fast-reflecting mirror execution module is used to drive the first fast-reflecting mirror and the second fast-reflecting mirror to deflect in coordination based on the output of the target four-channel piezoelectric driving voltage. The main control module is also used to control the spot detection module to perform the next closed-loop control cycle and re-acquire the second spot positions of the near-field detection surface and the far-field detection surface.

2. The system according to claim 1, characterized in that, The system also includes a host computer communication module, which is used to receive the target spot coordinates, matrix parameters, control parameters and scanning parameters sent by the host computer during initialization, and send them to the main control module.

3. The system according to claim 2, characterized in that, The main control module includes programmable logic and a processing system; The processing system is used to sequentially select each of the four piezoelectric drive channels as the current scanning channel, and control the drive output module to apply positive and negative stepped drive voltages to the current scanning channel based on the scanning parameters, so that the dual fast mirror execution module deflects the first fast mirror and the second fast mirror according to each drive voltage. The spot detection module is specifically used to acquire the first spot position of the near-field detection surface and the far-field detection surface after each driving voltage is applied, and send it to the main control module; The programmable logic is used to calculate the position of the first spot for each driving voltage to obtain multiple first spot coordinates, including near-field X-axis spot coordinates, near-field Y-axis spot coordinates, far-field X-axis spot coordinates and far-field Y-axis spot coordinates. The processing system is also used to calculate multiple spot coordinate changes based on the reference coordinates and multiple first spot coordinates for each driving voltage, and to calculate the driving voltage change based on the reference voltage and each driving voltage in the scanning parameters. The processing system is also used to perform linear fitting on the change in coordinates of each spot based on all sampling points to obtain multiple fitting slopes, which are combined into the coupling coefficient of the current scanning channel. All coupling coefficients of the four piezoelectric drive channels are combined into the coupling matrix. Each sampling point includes the change in drive voltage corresponding to a drive voltage and multiple changes in spot coordinates.

4. The system according to claim 3, characterized in that, Before performing linear fitting on the coordinate changes of each spot based on all sampling points, the processing system is further configured to: Based on all sampling points, linear fitting was performed on the coordinate changes of each spot to obtain multiple fitted straight lines; For each sampling point, calculate the residual between the sampling point and each fitted line; Calculate the mean and standard deviation based on the residuals of all sampling points; Based on the mean and the standard deviation, determine whether each sampling point is abnormal; Remove sampling points that are judged to be abnormal.

5. The system according to claim 3, characterized in that, The processing system is also used to perform a regularized pseudo-inverse on the coupling matrix to obtain a decoupling matrix.

6. The system according to claim 3, characterized in that, The programmable logic is also used to calculate the position of the second spot to obtain multiple second spot coordinates; The processing system is further configured to calculate the error vector based on the coordinates of the plurality of second light spots and the coordinates of the target light spot, and to calculate the control correction amount of the error vector based on the matrix parameters using a closed-loop control algorithm, and to perform matrix multiplication operation on the control correction amount and the decoupling matrix to obtain the four-channel piezoelectric drive voltage increment. The processing system is further configured to superimpose the incremental values ​​of the four piezoelectric drive voltages onto the four piezoelectric drive voltages output by the drive output module in the previous closed-loop control cycle to obtain the target four piezoelectric drive voltages.

7. The system according to claim 3, characterized in that, The processing system is also used to continuously determine whether the decoupling matrix has failed, and to regenerate the coupling matrix when the decoupling matrix fails.

8. The system according to claim 6, characterized in that, The drive output module includes an adjustment unit, a digital-to-analog conversion unit, and an amplification unit; The adjustment unit is used to smooth the target four-channel piezoelectric drive voltage; The digital-to-analog converter is used to convert the smoothed target four piezoelectric drive voltages into four analog voltage signals. The amplification unit is used to amplify the four analog voltage signals and output the amplified four analog voltage signals to the dual fast mirror execution module.

9. The system according to claim 8, characterized in that, The processing system is also used to continuously monitor the coordinates of the multiple second light spots and the four analog voltage signals before and after amplification based on the control parameters, and to perform corresponding protection operations when an anomaly is detected.

10. A method for stabilizing beam pointing in a dual-fast mirror system based on matrix decoupling, characterized in that, The method includes: The driving voltage change is obtained by sequentially scanning the four piezoelectric driving channels. The first spot position of the near-field and far-field detection surfaces is received after each scan. A coupling matrix is ​​generated based on the driving voltage change and the first spot position. The decoupling matrix of the coupling matrix is ​​calculated. The positions of the second light spots on the near-field detection surface and the far-field detection surface are acquired in real time. An error vector is constructed based on the position of the second spot. The control correction amount of the error vector is calculated, and the control correction amount is mapped to the four-channel piezoelectric drive voltage increment using the decoupling matrix to obtain the target four-channel piezoelectric drive voltage. The target four-channel piezoelectric drive voltage is output and processed; Based on the target four-channel piezoelectric driving voltage after output processing, the first fast mirror and the second fast mirror are driven to deflect in coordination. Proceed to the next closed-loop control cycle and return to the step of real-time acquisition of the second spot positions of the near-field detection surface and the far-field detection surface.

Images