Metal resonator gyroscope temperature drift online compensation system and method based on parameter real-time identification

Abstract

The invention relates to a metal resonator gyroscope temperature drift online compensation system and method based on parameter real-time identification. The system comprises a metal resonator gyroscope, a field programmable gate array (FPGA), a temperature sensor, an analog-to-digital converter (ADC), a digital-to-analog converter (DAC), a microcontroller ARM and a memory. The metal resonant gyroscope is used for sensing the external angular velocity; the field programmable gate array FPGA is used for generating an excitation signal and detecting a vibration signal; the temperature sensor is used for detecting the working temperature of the gyroscope in real time; the ADC is used for synchronously converting the collected gyroscope vibration signal, temperature signal and excitation signal; the digital-to-analog converter DAC is used for converting an excitation signal generated by the field programmable gate array FPGA and outputting the excitation signal to the metal resonant gyroscope; the microcontroller ARM is responsible for operating a parameter identification algorithm and compensation calculation; the memory is used for storing initial model parameters, algorithm programs and real-time data. According to the invention, high-precision compensation in a full-temperature zone can be realized, and the zero-bias stability, the scale factor linearity and the environmental adaptability of the gyroscope can be obviously improved.

Description

Technical Field

[0001] This invention belongs to the technical field of ship system technology - electronic information system - navigation system, and relates to metal resonant gyroscope compensation technology for suppressing temperature drift. Specifically, it relates to an online temperature drift compensation system and method for metal resonant gyroscopes based on real-time parameter identification. Background Technology

[0002] Metal resonant gyroscopes, due to their all-solid-state structure, high reliability, and long lifespan, have broad application prospects in aerospace, marine, and high-end industries. However, the physical properties (such as Young's modulus and coefficient of thermal expansion) of their core sensing element—the metal resonator—are significantly affected by temperature, causing drift in key parameters such as resonant frequency, quality factor (Q value), and modal symmetry. This physical phenomenon ultimately manifests as drastic changes in the gyroscope's output zero bias and scaling factor with temperature, becoming the primary bottleneck restricting its accuracy improvement.

[0003] Currently, the most commonly used temperature compensation techniques in industry are the "table lookup method" or the "static polynomial fitting method." This involves calibrating the gyroscope across the entire temperature range in a laboratory temperature test chamber to establish a one-dimensional or multi-dimensional static mapping table between the gyroscope output and temperature. In practical applications, temperature errors are compensated by interpolating from this table. However, this method has several inherent drawbacks: The calibration process is extensive and costly: each gyroscope requires temperature testing that can last for tens of hours.

[0004] Poor adaptability: The compensation model is static and cannot reflect the changes in the gyroscope's own performance over time (such as material fatigue and aging), resulting in poor long-term stability.

[0005] Individual variation problem: Even within the same batch of gyroscopes, their parameters vary, making static models unusable and requiring individual calibration, resulting in low production efficiency.

[0006] Unable to reflect thermal transient effects: Static models are usually collected at steady-state temperatures and have limited ability to suppress errors during rapid temperature changes. Even if the model is built in a dynamic temperature environment, its ability to suppress errors caused by temperature changes that differ from the modeling environment is still limited.

[0007] Therefore, developing a dynamic compensation method that can track the internal state of a gyroscope online, adapt to environmental changes, and eliminate the need for repeated calibration is crucial for unlocking the performance potential of metal resonant gyroscopes. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention proposes an online temperature drift compensation system and method for metal resonant gyroscopes based on real-time parameter identification. This system and method can achieve high-precision compensation across the entire temperature range and significantly improve the zero-bias stability, scaling factor linearity, and environmental adaptability of gyroscopes.

[0009] One of the above-mentioned objectives of the present invention is achieved by the following technical solution: An online temperature drift compensation system for a metal resonant gyroscope based on real-time parameter identification includes a metal resonant gyroscope, a field-programmable gate array (FPGA), a temperature sensor, an analog-to-digital converter (ADC), a digital-to-analog converter (DAC), an ARM microcontroller, and a memory. The metal resonant gyroscope is used to sense external angular velocity. The field-programmable gate array (FPGA) is connected to an analog-to-digital converter (ADC) and a digital-to-analog converter (DAC) for generating excitation signals and detecting vibration signals. The temperature sensor is attached to the resonator base and is used to detect the operating temperature of the gyroscope in real time; The analog-to-digital converter (ADC) is used to synchronously convert the acquired gyroscope vibration signal, temperature signal, and excitation signal; the digital-to-analog converter (DAC) is used to convert the excitation signal generated by the field-programmable gate array (FPGA) and output it to the metal resonant gyroscope. The microcontroller ARM is responsible for running the parameter identification algorithm and compensation calculation. The memory is connected to a field-programmable gate array (FPGA) and a microcontroller (ARM) and is used to store initial model parameters, algorithm programs, and real-time data.

[0010] Moreover, the field-programmable gate array (FPGA) consists of modules such as a phase-locked loop (PLL), a modem, and a direct digital frequency synthesizer (DDS).

[0011] Moreover, the temperature sensor uses a high-precision PT1000 platinum resistance thermometer.

[0012] Furthermore, both the analog-to-digital converter (ADC) and the digital-to-analog converter (DAC) employ converters with a bit width of no less than 16 bits and a sampling rate of no less than 200K.

[0013] Furthermore, the microcontroller ARM is an STMicroelectronics STM32H7 series microcontroller; the memory is preferably an off-chip FLASH.

[0014] Moreover, the real-time parameter identification algorithm adopts the recursive least squares method FFRLS with forgetting factor, using the gyroscope output signal as the observation vector to estimate the key parameters of the harmonic oscillator quality factor Q and vibration amplitude online in real time; the forgetting factor ranges from 0.95 to 0.999.

[0015] Furthermore, the compensation calculation includes establishing a multivariate error model based on internal parameters and performing real-time compensation. The multivariate error model includes a zero-bias model and a calibration factor model; the zero-bias model is shown in equation (1): (1) The calibration factor model is shown in equation (2): (2) The compensation model formula is shown in equation (3): (3) in, The zero bias value after compensation, The zero bias value before compensation, The zero-bias compensation value is calculated using formula (1). This is the scaling factor compensation coefficient.

[0016] The second objective of this invention is achieved through the following technical solution: A compensation method for an online temperature drift compensation system for a metal resonant gyroscope based on real-time parameter identification, comprising the following steps: Step 1, Initialization: Power on the system and load the program and initial model parameters from memory; Step 2: Data Acquisition: The Field Programmable Gate Array (FPGA) reads the gyroscope vibration signal through an analog-to-digital converter (ADC). , excitation signal and temperature information and transmit temperature information via communication bus gyroscope resonant frequency Phase difference Transmitted to the microcontroller; Step 3: Parameter Identification: The microcontroller calls the recursive least squares subroutine with a forgetting factor, based on the current gyroscope vibration signal. , excitation signal and phase difference The quality factor Q value at the current moment is estimated online. Step 4: Update the model: Add temperature information gyroscope resonant frequency Quality factor at the current moment Substitute the values ​​into the zero-biased model to update the model coefficients, so that the model dynamically approximates the current optimum; Step 5: Perform compensation calculations: Calculate the zero-bias compensation value and scale factor compensation coefficient based on the updated zero-bias error model and scaling error model; Step 6: Perform compensation output: Compensate the original output of the gyroscope according to the compensation model formula, send the compensated gyroscope output in the form of serial port, and return to step 2 to start the loop for the next moment.

[0017] The advantages and positive effects of this invention are as follows: 1. This invention overcomes the shortcomings of traditional static compensation methods in tracking the aging of gyroscopes over time by identifying internal parameters and updating the model in real time, and achieves dynamic adaptive compensation.

[0018] 2. The compensation method based on internal parameters of this invention is insensitive to individual differences in gyroscopes, reduces the workload of individual calibration, has strong robustness, and improves efficiency.

[0019] 3. The present invention employs a recursive least squares method with a forgetting factor, which has low computational complexity and high efficiency. It can achieve online real-time compensation at the millisecond or even micrometer level on microcontrollers with limited resources, thus meeting the real-time requirements of the control system.

[0020] 4. This invention can fundamentally improve the upper limit of gyroscope accuracy, enabling it to maintain excellent performance even in harsh temperature environments and broaden its application areas.

[0021] In summary, the method of the present invention can achieve high-precision compensation across the entire temperature range, and can significantly improve the zero-bias stability, scaling factor linearity, and environmental adaptability of the gyroscope. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the online temperature drift compensation system for a metal resonant gyroscope based on real-time parameter identification, as described in this invention. Figure 2 This is a flowchart of the online compensation system of the present invention; Figure 3 This is a schematic diagram illustrating the principle of online parameter identification using the recursive least squares method employed in this invention. Detailed Implementation

[0023] The structure of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0024] Please refer to the following for an online temperature drift compensation system for metal resonant gyroscopes based on real-time parameter identification: Figure 1 The invention features an online compensation system comprising a metal resonant gyroscope, a field-programmable gate array (FPGA), a temperature sensor, an analog-to-digital converter (ADC), a digital-to-analog converter (DAC), an ARM microcontroller, and a memory.

[0025] The metal resonant gyroscope is used to sense external angular velocity.

[0026] The field-programmable gate array (FPGA) consists of modules such as a phase-locked loop (PLL), a modem, and a direct digital frequency synthesizer (DDS). The PLL controls the phase difference between the gyroscope vibration signal and the excitation signal. The modem generates the excitation signal and decodes the gyroscope vibration signal, respectively. The DDS generates the reference signal required by the modulator. It is connected to an analog-to-digital converter (ADC) and a digital-to-analog converter (DAC) for generating the excitation signal and detecting the vibration signal.

[0027] The temperature sensor is mounted on the resonator base and is used to detect the gyroscope's operating temperature in real time. In this invention, a high-precision PT1000 platinum resistance thermometer is preferably used as the temperature sensor.

[0028] Both the analog-to-digital converter (ADC) and the digital-to-analog converter (DAC) employ converters with a bit width of at least 16 bits and a sampling rate of at least 200K. The ADC is used to synchronously convert the acquired gyroscope vibration signal, temperature signal, and excitation signal. The DAC is used to convert the excitation signal generated by the field-programmable gate array (FPGA) and output it to the metal resonant gyroscope.

[0029] The microcontroller ARM is responsible for running the parameter identification algorithm and compensation calculation. In this invention, the microcontroller ARM is preferably an STMicroelectronics STM32H7 series microcontroller.

[0030] The memory, connected to a Field Programmable Gate Array (FPGA) and an ARM microcontroller, is used to store initial model parameters, algorithm programs, and real-time data. In this invention, the memory is preferably an off-chip FLASH memory.

[0031] The core of the software algorithm is: 1) Real-time parameter identification algorithm: The recursive least squares method with forgetting factor FFRLS is adopted. The gyroscope output signal (driving amplitude, detection amplitude, phase difference) is used as the observation vector to estimate key parameters such as the quality factor Q and vibration amplitude of the harmonic oscillator in real time online. The forgetting factor ranges from 0.95 to 0.999 to ensure that the algorithm is more sensitive to the current data than to the historical data.

[0032] 2) Dynamic error modeling: Establishing a multivariate error model based on intrinsic parameters: The zero-biased model is shown in equation (1): (1) The calibration factor model is shown in equation (2): (2) in, For temperature, Let Q be the resonant frequency of the harmonic oscillator, and let Q be the quality factor of the harmonic oscillator. The scaling factor of the gyroscope, the coefficient The model is dynamically adapted to changes in the gyroscope state by updating in real time using a recursive least squares algorithm.

[0033] 3) Real-time compensation: The gyroscope output signal is directly compensated by the microcontroller, and the compensation model formula is shown in equation (3): (3) in, The zero bias value after compensation, The zero bias value before compensation, The zero-bias compensation value is calculated using formula (1). This is the scaling factor compensation coefficient.

[0034] For the compensation method of the above-mentioned online temperature drift compensation system for metal resonant gyroscopes based on real-time parameter identification, please refer to [link to relevant documentation]. Figure 2 and Figure 3 It includes the following steps: Step 1, Initialization: Power on the system and load the program and initial model parameters from memory; Step 2: Data Acquisition: The Field Programmable Gate Array (FPGA) reads the gyroscope vibration signal through an analog-to-digital converter (ADC). , excitation signal and temperature information and transmit temperature information via communication bus gyroscope resonant frequency Phase difference Transmitted to the microcontroller; Step 3: Parameter Identification: The microcontroller calls the recursive least squares subroutine with a forgetting factor, based on the current gyroscope vibration signal. , excitation signal and phase difference The quality factor Q value at the current moment is estimated online. Step 4: Update the model: Add temperature information gyroscope resonant frequency Quality factor at the current moment Substitute the values ​​into the zero-biased model to update the model coefficients, so that the model dynamically approximates the current optimum; Step 5: Perform compensation calculations: Calculate the zero-bias compensation value and scale factor compensation coefficient based on the updated zero-bias error model and scaling error model; Step 6: Perform compensation output: Compensate the original output of the gyroscope according to the compensation model formula, send the compensated gyroscope output in the form of serial port, and return to step 2 to start the loop for the next moment.

[0035] Although embodiments and drawings of the present invention have been disclosed for illustrative purposes, those skilled in the art will understand that various substitutions, variations and modifications are possible without departing from the spirit and scope of the present invention and the appended claims. Therefore, the scope of the present invention is not limited to the contents disclosed in the embodiments and drawings.

Claims

1. An online temperature drift compensation system for a metal resonant gyroscope based on real-time parameter identification, characterized in that: This includes metal resonant gyroscopes, field-programmable gate arrays (FPGAs), temperature sensors, analog-to-digital converters (ADCs), digital-to-analog converters (DACs), ARM microcontrollers, and memory; The metal resonant gyroscope is used to sense external angular velocity. The field-programmable gate array (FPGA) is connected to an analog-to-digital converter (ADC) and a digital-to-analog converter (DAC) for generating excitation signals and detecting vibration signals. The temperature sensor is attached to the resonator base and is used to detect the operating temperature of the gyroscope in real time; The analog-to-digital converter (ADC) is used to synchronously convert the acquired gyroscope vibration signal, temperature signal, and excitation signal. The digital-to-analog converter (DAC) is used to convert the excitation signal generated by the field-programmable gate array (FPGA) and output it to the metal resonant gyroscope. The microcontroller ARM is responsible for running the parameter identification algorithm and compensation calculation. The memory is connected to a field-programmable gate array (FPGA) and a microcontroller (ARM) and is used to store initial model parameters, algorithm programs, and real-time data.

2. The online temperature drift compensation system for a metal resonant gyroscope based on real-time parameter identification according to claim 1, characterized in that: The field-programmable gate array (FPGA) consists of modules such as a phase-locked loop (PLL), a modem, and a direct digital frequency synthesizer (DDS).

3. The online temperature drift compensation system for a metal resonant gyroscope based on real-time parameter identification according to claim 1, characterized in that: The temperature sensor uses a high-precision PT1000 platinum resistance thermometer.

4. The online temperature drift compensation system for a metal resonant gyroscope based on real-time parameter identification according to claim 1, characterized in that, Both the analog-to-digital converter (ADC) and the digital-to-analog converter (DAC) use converters with a bit width of not less than 16 bits and a sampling rate of not less than 200K.

5. The online temperature drift compensation system for a metal resonant gyroscope based on real-time parameter identification according to claim 1, characterized in that, The microcontroller ARM is an STMicroelectronics STM32H7 series microcontroller; the memory is preferably an off-chip FLASH.

6. The online temperature drift compensation system for a metal resonant gyroscope based on real-time parameter identification according to claim 1, characterized in that: The real-time parameter identification algorithm adopts the recursive least squares method with forgetting factor (FFRLS), using the gyroscope output signal as the observation vector to estimate the key parameters of the harmonic oscillator quality factor Q and vibration amplitude online in real time; the forgetting factor ranges from 0.95 to 0.

999.

7. The online temperature drift compensation system for a metal resonant gyroscope based on real-time parameter identification according to claim 6, characterized in that: The compensation calculation includes establishing a multivariate error model based on internal parameters and performing real-time compensation. The multivariate error model includes a zero-bias model and a calibration factor model; the zero-bias model is shown in equation (1): （1） The calibration factor model is shown in equation (2): （2） The compensation model formula is shown in equation (3): （3） in, The zero bias value after compensation, The zero bias value before compensation, The zero-bias compensation value is calculated using formula (1). This is the scaling factor compensation coefficient.

8. A compensation method for an online temperature drift compensation system for a metal resonant gyroscope based on real-time parameter identification as described in claim 7, characterized in that: Includes the following steps: Step 1, Initialization: Power on the system and load the program and initial model parameters from memory; Step 2: Data Acquisition: The Field Programmable Gate Array (FPGA) reads the gyroscope vibration signal through an analog-to-digital converter (ADC). , excitation signal and temperature information and transmit temperature information via communication bus gyroscope resonant frequency Phase difference Transmitted to the microcontroller; Step 3: Parameter Identification: The microcontroller calls the recursive least squares subroutine with a forgetting factor, based on the current gyroscope vibration signal. , excitation signal and phase difference The quality factor Q value at the current moment is estimated online. Step 4: Update the model: Add temperature information gyroscope resonant frequency Quality factor at the current moment Substitute the values ​​into the zero-biased model and the zero-biased model to update the model coefficients, so that the model dynamically approximates the current optimum; Step 5: Perform compensation calculations: Calculate the zero-bias compensation value and scale factor compensation coefficient based on the updated zero-bias error model and scaling error model; Step 6: Perform compensation output: Compensate the original output of the gyroscope according to the compensation model formula, send the compensated gyroscope output in the form of serial port, and return to step 2 to start the loop for the next moment.

Images