A four-point piezoelectric driven fast-swing mirror control system based on STM32

Abstract

This invention discloses a four-point piezoelectric driven fast-swing mirror control system based on STM32, including a piezoelectric driven fast-swing mirror, a PSD position-sensitive detector, an AD acquisition module, a DA output module, a piezoelectric drive controller, piezoelectric actuators, an STM32 control module, and the PSD position-sensitive detector. The electrical signal of the object to be tracked is transmitted back to the STM32 controller via the AD acquisition module. An improved ADRC algorithm is used to calculate the driving voltage of each piezoelectric ceramic actuator. Subsequently, the STM32 controller sends control signals through the DA output module, which are then driven by the piezoelectric ceramic actuators to control the rotation angle of the fast-swing mirror. This invention fully utilizes the parallel computing advantages of STM32, reducing system latency and achieving high-precision, real-time, and synchronous multi-piezoelectric compensation and control. It is of great significance for closed-loop testing of the fast-swing mirror controller, piezoelectric ceramic hysteresis nonlinear modeling, and model parameter identification research.

Description

Technical Field

[0001] This invention relates to a two-dimensional fast-swing mirror with a piezoelectric ceramic driving element, and more specifically to a piezoelectric driven fast-swing mirror control system based on STM32. Background Technology

[0002] Fast Steering Mirrors (FSMs) are used for high-precision, high-dynamic pointing of light beams. As a crucial component of systems such as photoelectric tracking, they offer advantages such as fast response, high resolution, high accuracy, and high bandwidth, and are widely used in fields such as line-of-sight stabilization and image shift compensation, optical scanning, photoelectric stabilization tracking, laser weapons, and space laser communication. FSMs are mainly classified into two types based on their driving methods: piezoelectric ceramic driven and voice coil motor driven. They employ resistance strain gauge displacement sensors and capacitive displacement sensors to measure the deformation of the piezoelectric ceramic and the oscillation of the fast steering mirror, and utilize closed-loop control to achieve rapid and accurate positioning.

[0003] Piezoelectric ceramic driven FSMs (Frequency Swing Mirrors) feature high output torque and wide response bandwidth, making them suitable for applications with high bandwidth requirements. They are commonly used as drive devices for fast-swinging mirror actuators in image stabilization systems, offering higher resonant frequencies and displacement resolution compared to voice coil motors (VCAs). However, this type of FSM has a small total stroke, high driving voltage, and relatively complex driving circuitry. Furthermore, piezoelectric ceramics themselves exhibit nonlinear characteristics such as hysteresis and creep, necessitating compensation through software algorithms.

[0004] Currently, the most common control method for fast-swinging mirrors is PID control. This control method is simple in principle, relatively easy to implement in engineering, and can be applied to most scenarios. However, PID control also has many problems. Since the fast-swinging mirror in this patent is used for optical path stabilization, it requires high control accuracy and is a nonlinear system; therefore, traditional PID linear control is difficult to meet the actual working requirements.

[0005] The STM32 is a powerful microcontroller series designed by STMicroelectronics based on the Cortex-M core. While possessing powerful functionality, the STM32 series also offers low cost, high performance, and ease of operation. Furthermore, it boasts abundant I / O resources, allowing for convenient connection with other modules to achieve rapid communication. Users can reconfigure the STM32's I / O resources and select appropriate peripherals such as IIC and SPI according to their needs, thus saving the cost of purchasing additional external devices. With a mains frequency of up to 72MHz, it can stably execute high-speed and complex hysteresis compensation calculations, making it a highly cost-effective microcontroller series. Therefore, the STM32 was used to implement this piezoelectric driven fast-swing mirror control system.

[0006] Active disturbance rejection control (ADRC) technology was proposed by researcher Han Jingqing in 1980. It is a control strategy that has very little dependence on the mathematical model of the controlled object. It uses feedback control technology to estimate and compensate for disturbances inside and outside the system in real time, so as to achieve precise control of the motor output and has strong robustness.

[0007] Sliding Mode Control (SMC) is a control theory built upon modern control theory. The core of sliding mode control is to establish a sliding surface, pull the controlled system onto this surface, and make the system move along it. The advantage of sliding mode control is that it ignores external disturbances and uncertainties, employing a relatively forceful method to achieve the control objective. Summary of the Invention

[0008] To address the aforementioned problems in existing technologies, this invention proposes a four-point piezoelectric driven fast-swing mirror control system based on STM32.

[0009] The present invention adopts the following technical solution:

[0010] A four-point piezoelectric driven fast-swing mirror control system based on STM32 includes a piezoelectric driven fast-swing mirror, a PSD position-sensitive detector, an AD acquisition module, a DA output module, a piezoelectric ceramic driver, a piezoelectric ceramic actuator, and an STM32 control module. The PSD position-sensitive detector transmits the electrical signal of the object to be tracked back to the STM32 controller through the AD acquisition module. The driving voltage of each piezoelectric ceramic actuator is obtained through an improved ADRC algorithm. Then, the STM32 controller sends a control signal through the DA output module, which drives the piezoelectric ceramic driver to complete the control of the rotation angle of the fast-swing mirror.

[0011] Furthermore, the improved ADRC algorithm employs a linear state error feedback control system based on sliding mode control; the expression for the linear state error feedback control system based on sliding mode control in ADRC is:

[0012]

[0013] Where e represents the system output error, v represents the reference input generated by the tracking differentiator (TD), and z represents the system state estimated by the extended observer (ESO). Represents the target value of the control quantity; and k are the gain coefficients; c are the sliding surface parameters; For disturbance compensation, representing the characteristics of the controlled object, it can be derived from the initial acceleration of the step object. The larger the value, the weaker the anti-interference capability. u represents the output that directly affects the controlled object.

[0014] Furthermore, the STM32 control module adopts an improved ADRC algorithm, which tracks the target signal by tracking the transition signal output by the differentiator. The differential signal of the transition signal output by the differentiator is used as the target differential signal. At the same time, the differential signal of the feedback state output by the extended state observer (ESO) is introduced as the actual differential signal, thereby achieving effective tracking of the differentiator, avoiding jitter problems, and improving the stability and robustness of the system.

[0015] Furthermore, the piezoelectric-driven fast-swing mirror uses piezoelectric actuators as drivers and drives the mirror surface through a flexible support micro-displacement amplification mechanism. The two-dimensional deflection of the mirror surface adopts a four-point parallel driving method, with the four piezoelectric actuators arranged in a rectangle around the center of the mirror surface.

[0016] Furthermore, the piezoelectric drive controller is connected to the STM32 control module. The piezoelectric drive controller includes a power supply module, a DA module, an AD module, a sensing module, a servo module, and a drive module. The drive module includes four independent drive channels for parallel driving of the four-point piezoelectric ceramic actuator.

[0017] Furthermore, the AD module input is 0-10V, the sampling rate is not less than 20kHz, and the DA module output is 0-10V.

[0018] Furthermore, the piezoelectric drive controller is connected to the DA output module, and a drive circuit is used to amplify the output voltage of the DA output module, amplifying the 0-10V analog signal into a 0-120V high voltage output, thereby realizing the drive of the piezoelectric drive fast swing mirror.

[0019] Furthermore, the PSD position-sensitive detector is connected to the STM32 control module for spot centroid acquisition, and helps the fast-swinging mirror point to the correct position by converting the optical signal into an electrical signal.

[0020] Furthermore, the piezoelectric actuator is connected to a piezoelectric driven fast swing mirror, with a driving voltage of 0-150V, using piezoelectric ceramic as the basic element, and outputting displacement and force.

[0021] Furthermore, the piezoelectric-driven fast-swing mirror is controlled by four piezoelectric ceramic actuators. When the mirror needs to rotate along the Y-axis, one set of piezoelectric actuators on the same side remains in place, while the other set of piezoelectric actuators on the same side actuates. When the mirror needs to rotate along the X-axis, one set of diagonally opposite piezoelectric actuators remains in place, while the other set of diagonally opposite piezoelectric actuators actuates.

[0022] The present invention has the following beneficial effects:

[0023] This invention fully utilizes the parallel computing advantages of STM32, reduces system computation latency, improves system response speed, and achieves high-precision, real-time, and synchronous multi-piezoelectric compensation and control. It is of great significance for closed-loop testing of fast-swing mirror controllers, piezoelectric ceramic hysteresis nonlinear modeling, and model parameter identification. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the working principle of a four-point piezoelectric driven fast-swing mirror mechanism.

[0025] Figure 2 This is a hardware connection diagram of a four-point piezoelectric driven fast-swing mirror control system.

[0026] Figure 3 This is a block diagram of the control system for a four-point piezoelectric-driven fast-swing mirror.

[0027] Figure 4 This is a block diagram of the ADRC algorithm model structure.

[0028] Figure 5 This is a block diagram of the improved sliding mode ADRC control principle. Detailed Implementation

[0029] The present invention will now be described in further detail with reference to the accompanying drawings.

[0030] This invention provides a high-precision, fast-response piezoelectric-driven fast-swing mirror control system based on STM32. The system includes a piezoelectric-driven fast-swing mirror, a PSD position-sensitive detector, an AD acquisition module, a DA output module, a piezoelectric drive controller, a piezoelectric actuator, and an STM32 control module. The hardware connections are as follows: Figure 2 As shown, the system structure is as follows Figure 3 As shown, the PSD position-sensitive detector transmits the electrical signal of the light spot to be tracked back to the STM32 controller through the AD acquisition module. The improved ADRC algorithm is used to obtain the driving voltage of each piezoelectric ceramic actuator. Then, the STM32 controller sends a control signal through the DA output module, which is driven by the piezoelectric ceramic driver to complete the control of the rotation angle of the fast swing mirror.

[0031] In the improved ADRC algorithm designed within the STM32 control module, a tracking differentiator (TD) outputs a transition signal to track the target signal, avoiding large initial errors. The tracking differentiator (TD) can output the differential signal of the transition signal as the target differential signal, while an extended state observer (ESO) is introduced to output the differential signal of the feedback state as the actual differential signal. This achieves effective tracking of the differentiator, avoids jitter problems, and thus improves the stability and robustness of the system.

[0032] The improved ADRC algorithm addresses the issues of weak anti-interference capability and poor control performance of PD control in the linear state error feedback (LSEF) control law of the traditional ADRC algorithm by adopting a sliding mode control-linear state error feedback (SMC-LSEF) system.

[0033] Preferably, the four-point piezoelectric driven fast swing mirror uses piezoelectric ceramic actuators as actuators and drives the mirror surface through a flexible support micro-displacement amplification mechanism. The two-dimensional deflection of the mirror surface adopts a four-point parallel driving method, with the four piezoelectric ceramic actuators arranged in a rectangle around the center of the mirror surface.

[0034] Preferably, the piezoelectric actuator is connected to the STM32 control module. The piezoelectric actuator includes a power supply module, a DA module, an AD module, a sensing module, a servo module, and a drive module. The drive module contains four independent drive channels for parallel driving of the four-point piezoelectric ceramic actuator. The AD module has an input of 0–10V and a sampling rate of not less than 20kHz. The DA module has an output of 0–10V.

[0035] Preferably, the piezoelectric driver is connected to the 16-bit DAC, and a driving circuit amplifies the output voltage of the DAC, increasing the 0-10V analog signal to a 0-120V high-voltage output, thereby driving the piezoelectric-driven fast-swing mirror. The piezoelectric actuator is connected to the piezoelectric-driven fast-swing mirror, with a driving voltage of 0-150V, using piezoelectric ceramic as the basic element, and outputs displacement and force.

[0036] Preferably, the position-sensitive detector (PSD) is connected to the STM32 control module for spot centroid acquisition, and helps the fast-swinging mirror point to the correct position by converting the optical signal into an electrical signal.

[0037] Preferably, the ADRC algorithm designed within the STM32 control module uses a tracking differentiator (TD) to output a transition signal to track the target signal, avoiding large initial errors. The tracking differentiator (TD) can output the differential signal of the transition signal as the target differential signal, while simultaneously introducing an extended state observer (ESO) to output the differential signal of the feedback state as the actual differential signal. This achieves effective tracking of the differentiator, avoids jitter problems, and thus improves the stability and robustness of the system.

[0038] Preferably, the piezoelectrically driven fast-swivel mirror is controlled by four piezoelectric ceramic actuators. When the mirror needs to rotate along the Y-axis, actuators 1 and 4 remain stationary, while actuators 2 and 3 actuate; similarly, when the mirror needs to rotate along the X-axis, actuators 1 and 2 remain stationary, while actuators 3 and 4 actuate. Figure 1 As shown.

[0039] Preferably, to address the issues of weak anti-interference capability and poor control performance of PD control in the linear state error feedback (LSEF) control law of the traditional ADRC algorithm, a sliding mode control-linear state error feedback (SMC-LSEF) system is adopted.

[0040] This invention employs the ADRC algorithm to overcome the problems of slow error response and inaccurate compensation often caused by traditional PID algorithms. Traditional PID algorithms only compensate for control after an error occurs, while ADRC compensates for observed disturbances at the output immediately. The output is then transmitted through a 16-bit DA module circuit, which drives four piezoelectric ceramic actuators via a piezoelectric ceramic driver, thereby improving the system's stability and robustness. Figure 4 As shown.

[0041] Traditional PID algorithms often struggle to balance speed and overshoot when dealing with step responses or large initial errors. The ADRC algorithm, however, introduces a tracking differentiator (TD) to output a transition signal that tracks the target signal, thus avoiding large initial errors. While the D-term differential signal in traditional PID algorithms has predictive capabilities, it merely subtracts the previous error from the current error, resulting in a coarse differential result that is difficult to obtain in practice. The tracking differentiator, on the other hand, can output the differential signal of the transition signal as the target differential signal, and simultaneously introduces an extended state observer (ESO) to output the differential signal of the feedback state as the actual differential signal. The difference between these signals represents the error of the ADRC's differential action term. The discrete mathematical form of the tracking differentiator is shown in equation (1).

[0042] x1(k+1) = x1(k) + T* x2(k)

[0043] x2(k+1)= x2(k)+T* fhan(x1(k)-v,x2(k),r,h) (1)

[0044] Among them, T is the integration step size and the sampling time. fhan is a non-linear function, also known as the fast control optimal synthesis function, with inputs e = x1 - v, x2, the speed factor r, and the filtering factor h. For this function, the larger r is, the faster the tracking speed; the smaller h is, the better the noise filtering effect.

[0045] The extended state observer can expand the uncertain factors in the system into new state variables and estimate them. The ESO does not require an accurate mathematical model of the controlled object and can well estimate the internal state and uncertain factors of the system based on the input-output information. Its mathematical expression is as follows:

[0046] ε1 = z1(k) - y(k)

[0047] z1(k + 1) = z1(k) + h[z2(k) - β01ε1]

[0048] z2(k + 1) = z2(k) + h[z3(k) - β02fal(ε1, 1 / 2, δ) + bu]

[0049] z3(k + 1) = z3(k) + hβ02fal(ε1, 1 / 4, δ)

[0050] Among them, z is the real-time estimated quantity of the internal state variable of the system, and the parameter β is determined by the sampling time and order of the controlled object. The fal(ε1, a, δ) function is a continuous power function, e is the system error, a is the exponent, and δ is the boundary for distinguishing the magnitude of the error e.

[0051] The traditional PID control method generates the control quantity based on the "linear weighted sum" of the error ratio, differential, and integral. Although this method is simple, it may not be the best. Therefore, in ADRC, a non-linear combination method of error state feedback (NLESF) is tried, and the result significantly improves the processing efficiency of system information, and the non-linear feedback coefficient has a wider adaptability. Its mathematical expression is:

[0052] en = vn - zn

[0053] u0 = k1 fal(e1, a1, δ) + k2 fal(e2, a2, δ)

[0054] Among them, a and δ are adjustable parameters. Generally, a1 < 0 < a2 < 1, and δ is generally taken as 0.1 according to engineering experience.

[0055] In addition, in the control quantity u0, the system uncertain factor zn + 1 is compensated, and the non-linear system is converted into an integrator series type system, thus making the control problem of complex objects simpler. The mathematical expression of disturbance compensation is:

[0056]

[0057] Sliding mode control mainly includes the selection of the sliding mode switching surface s(x) and the design of the sliding mode reaching law u(x). The principle of the improved sliding mode ADRC control is as follows: Figure 5 As shown. For a nonlinear system, the state space is defined as: = f(x),x∈ There is a hyperplane in the state space: s(x) = s(x1, x2, ..., xn) = 0, which divides the state space into two parts: s > 0 and s < 0. Points on the switching surface are divided into three types: ordinary point C, starting point B, and ending point A. When both points approach the switching surface, the system's trajectory will move along s(x) = 0 to the ideal state. Therefore, the conditional expression that the system's motion satisfies in the sliding mode dynamic zone can be obtained as follows:

[0058]

[0059] The sliding surface function is determined as follows:

[0060]

[0061] Constructing a suitable sliding mode control law:

[0062]

[0063] In the formula ≠

[0064] As can be seen from the above, the realization of sliding mode motion requires the following three basic conditions to be met: (1) the existence of a sliding mode; (2) the satisfaction of the reachability condition. (3) Meets stability requirements.

[0065] Based on sliding mode control theory, a sliding surface is first constructed for the LESF system. Considering the suppression of high-frequency noise and the elimination of steady-state error, an integral quantity is introduced. Therefore, the integral sliding surface is defined as:

[0066] s= +c∫ dt

[0067] In the formula, c is the sliding surface parameter, and c>0.

[0068] To reduce the motion time of the moving point within the sliding mode dynamic zone and ensure a smooth transition process, an exponential reaching law is chosen as the sliding mode reaching law of the system, and its form is as follows:

[0069]

[0070] In the formula: >0, k>0 and both can be adjusted; sign() is the sign function.

[0071] Based on the Lyapunov stability principle, the stability of the sliding mode control system is determined. Given the selected sliding surface, the Lyapunov function is chosen as follows:

[0072]

[0073] Differentiating the above equation and combining it with the sliding mode convergence rate, we get:

[0074]

[0075] because Both k and are greater than 0, therefore The condition < 0 holds true. According to the Lyapunov stability criterion, the sliding mode motion tends to stabilize after the system's state point reaches the sliding mode dynamic region. Therefore, the expression for the improved LSEF based on sliding mode control in ADRC is:

[0076]

[0077] In summary, this invention discloses a four-point piezoelectric ceramic driven fast-swing mirror control system based on STM32, including an STM32 control module, a PSD position-sensitive detector, an AD acquisition module, a DA output module, a piezoelectric drive controller, a piezoelectric actuator, and a piezoelectric driven fast-swing mirror, wherein the piezoelectric driven fast-swing mirror is the controlled object. This system fully utilizes the parallel processing capabilities of STM32 to achieve control of a large-aperture fast-swing mirror based on an improved ADRC algorithm. Voltage data returned by the PSD position-sensitive detector is acquired through the AD acquisition circuit. The extension / retraction of the piezoelectric ceramic actuator is calculated using the improved ADRC algorithm designed within the system model. Finally, the data is output to the STM32 control module through the DA output module, thereby controlling the piezoelectric actuator to drive the fast-swing mirror and completing closed-loop control. This invention achieves closed-loop debugging of the controller's closed-loop control link and control algorithm, and is of great significance for closed-loop testing of the fast-swing mirror controller, piezoelectric ceramic hysteresis nonlinear modeling, and model parameter identification research.

[0078] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A four-point piezoelectric driven fast-swing mirror control system based on STM32, comprising a piezoelectric driven fast-swing mirror, a PSD position-sensitive detector, an AD acquisition module, a DA output module, a piezoelectric drive controller, a piezoelectric ceramic actuator, and an STM32 control module; characterized in that, The PSD position-sensitive detector transmits the electrical signal of the object to be tracked back to the STM32 controller through the AD acquisition module. The improved ADRC algorithm is used to obtain the driving voltage of each piezoelectric ceramic actuator. Then, the STM32 controller sends out control signals through the DA output module, which are driven by the piezoelectric drive controller to complete the control of the rotation angle of the fast swing mirror. The improved ADRC algorithm employs a linear state error feedback control system based on sliding mode control; the expression for the linear state error feedback control system based on sliding mode control in ADRC is: Where e represents the system output error, v represents the reference input generated by the tracking differentiator (TD), z represents the system state estimated by the extended observer (ESO), u0 represents the target value of the control variable, ξ and k are the gain coefficients, c is the sliding surface parameter, b0 is the disturbance compensation, which represents the characteristics of the controlled object and can be derived from the initial acceleration of the step object. The larger b0 is, the weaker the anti-disturbance capability; u is the output that directly acts on the controlled object.

2. The four-point piezoelectric driven fast-swing mirror control system based on STM32 according to claim 1, characterized in that, The STM32 control module employs an improved ADRC algorithm, which tracks the target signal by tracking the transition signal output by the differentiator. The differential signal of the transition signal output by the differentiator is used as the target differential signal. At the same time, the differential signal of the feedback state output by the extended state observer (ESO) is introduced as the actual differential signal, thereby achieving effective tracking of the differentiator, avoiding jitter problems, and improving the stability and robustness of the system.

3. The four-point piezoelectric driven fast-swing mirror control system based on STM32 according to claim 1, characterized in that, The piezoelectric-driven fast-swing mirror uses piezoelectric ceramic actuators as drivers and drives the mirror surface through a flexible support micro-displacement amplification mechanism. The two-dimensional deflection of the mirror surface adopts a four-point parallel driving method, with the four piezoelectric ceramic actuators arranged in a rectangle around the center of the mirror surface.

4. The four-point piezoelectric driven fast-swing mirror control system based on STM32 according to claim 1, characterized in that, The piezoelectric drive controller is connected to the STM32 control module. The piezoelectric drive controller includes a power supply module, a DA module, an AD module, a sensing module, a servo module, and a drive module. The drive module includes four independent drive channels for parallel driving of the four-point piezoelectric ceramic actuator.

5. The four-point piezoelectric driven fast-swing mirror control system based on STM32 according to claim 4, characterized in that, The AD module has an input of 0-10V and a sampling rate of not less than 20kHz, while the DA module has an output of 0-10V.

6. The four-point piezoelectric driven fast-swing mirror control system based on STM32 according to claim 1, characterized in that, The piezoelectric drive controller is connected to the DA output module. The drive circuit amplifies the output voltage of the DA output module, turning the 0-10V analog signal into a 0-120V high voltage output, thereby realizing the drive of the piezoelectric drive fast swing mirror.

7. The four-point piezoelectric driven fast-swing mirror control system based on STM32 according to claim 1, characterized in that, The PSD position-sensitive detector is connected to the STM32 control module and is used for spot centroid acquisition. It helps the fast-swinging mirror point to the correct position by converting the optical signal into an electrical signal.

8. The four-point piezoelectric driven fast-swing mirror control system based on STM32 according to claim 1, characterized in that, The piezoelectric ceramic actuator is connected to a piezoelectric driven fast swing mirror, with a driving voltage of 0-150V. It uses piezoelectric ceramic as the basic element and outputs displacement and force.

9. The four-point piezoelectric driven fast-swing mirror control system based on STM32 according to claim 1, characterized in that, The piezoelectric-driven fast-swing mirror is controlled by four piezoelectric ceramic actuators. When the mirror needs to rotate along the Y-axis, one set of piezoelectric ceramic actuators on the same side remains in place, while the other set of piezoelectric ceramic actuators on the same side actuates. When the mirror needs to rotate along the X-axis, one set of diagonally opposite piezoelectric ceramic actuators remains in place, while the other set of diagonally opposite piezoelectric ceramic actuators actuates.

Images