A Closed-Loop Control Method for the Longitudinal Axis Magnetic Field of a SERF Atomic Spin Gyroscope

Abstract

A closed-loop control method for the longitudinal magnetic field of a SERF atomic spin gyroscope is proposed. A sinusoidal modulated magnetic field, several times the electron resonance frequency, is applied along the x-axis. The system output signal is split into two paths: one path is input to a low-pass filter for the angular velocity response signal, and the other path is connected to a lock-in amplifier for phase demodulation. The demodulation result is fed back to the controller, thus achieving closed-loop control of the longitudinal magnetic field. This method achieves both decoupled measurement of angular velocity and longitudinal magnetic field and in-situ measurement of the longitudinal magnetic field. By establishing a mapping relationship with the longitudinal magnetic field through phase demodulation, it is theoretically unaffected by pump laser power, detection laser power, temperature, or other parameters. Furthermore, when the applied modulation frequency is higher than the electron resonance frequency, the electron spin exchange relaxation rate caused by the modulated magnetic field can be reduced, ensuring that the state is identical when the modulated magnetic field is applied to the unmodulated state. This also significantly improves the repeatability, nonlinearity, and zero-bias stability of the gyroscope's scaling factor.

Description

Technical Field

[0001] This invention relates to a closed-loop control method for the longitudinal axis magnetic field of a SERF atomic spin gyroscope, belonging to the field of gyroscope parameter closed-loop control technology. Background Technology

[0002] Atomic spin gyroscopes based on spin-exchange relaxation-free (SERF) exhibit extremely high sensitivity in theoretical angular velocity measurement, with a theoretical accuracy reaching 10⁻⁶. -6 With a speed on the order of ° / h and a theoretical limit of accuracy far exceeding that of other types of gyroscopes, it represents one of the important development directions for ultra-high precision inertial navigation for next-generation motion carriers.

[0003] From the perspective of the SERF atomic spin gyroscope principle, both magnetic fields and inertial inputs affect the gyroscope's output response. Therefore, factors such as changes in the environmental magnetic field can cause magnetic field drift sensed by the atoms, ultimately affecting the gyroscope's output stability. Specifically, magnetic field drift along the longitudinal axis not only exacerbates the electron spin exchange relaxation rate but also directly reduces the repeatability of the gyroscope's scaling factor, thus ultimately worsening the gyroscope's zero-bias stability. This significantly limits the improvement of SERF atomic spin gyroscope performance. For SERF atomic spin gyroscopes that prioritize long-term stability, achieving in-situ measurement of the magnetic field and suppressing magnetic field drift through closed-loop control has become an urgent problem to be solved. Summary of the Invention

[0004] The technical problem solved by this invention is: addressing the longitudinal axis (z-axis) magnetic field drift issue in SERF atomic spin inertial measurement systems, a closed-loop control method for the longitudinal axis magnetic field of a SERF atomic gyroscope is provided. A sinusoidal modulated magnetic field several times the electron resonance frequency is applied along the transverse axis (x-axis). Angular velocity can be measured using the DC quantity of the detected signal. Simultaneously, phase demodulation of the system's detected signal using a lock-in amplifier enables in-situ measurement of remanence along the longitudinal axis magnetic field. The phase demodulation result of the lock-in amplifier is fed back to the controller, thereby achieving closed-loop control of the longitudinal axis remanence. The demodulation phase is theoretically unaffected by pump laser power, detection laser power, temperature, etc. Furthermore, when the applied modulation frequency is higher than the electron resonance frequency, the electron spin exchange relaxation rate caused by the modulated magnetic field can be reduced, thus ensuring that the state under modulated magnetic field application is identical to the unmodulated state.

[0005] The technical solution of the present invention is as follows:

[0006] A closed-loop control method for the longitudinal magnetic field of a SERF atomic gyroscope is characterized by the following steps: After achieving polarization stabilization of atoms in the alkali metal gas cell of the SERF gyroscope system, a modulation magnetic field n times the electron resonance frequency is applied to the x-axis coil to avoid the additional electron spin exchange relaxation rate introduced by the applied modulation magnetic field, thereby ensuring that the scale factor in the closed-loop control state of the gyroscope is the same as that in the open-loop state, where n is an integer greater than or equal to 2; then, the system output signal is split into two paths to achieve in-situ decoupled measurement of angular velocity and magnetic field. One path is input to a low-pass filter, and after filtering, a DC component is output, which is used to output the angular velocity response signal. The other path is connected to a lock-in amplifier for phase demodulation, and the demodulated phase value reflects the magnitude of the longitudinal magnetic field. The phase demodulation result of the lock-in amplifier is fed back to the controller, thereby realizing closed-loop control of the Z-axis magnetic field.

[0007] This includes establishing a mapping relationship between the longitudinal magnetic field and the phase. The phase is theoretically unaffected by the pump laser power, the detection laser power, and the temperature, thereby avoiding the impact of other parameter drifts on the control effect.

[0008] The closed-loop control of the Z-axis magnetic field can reduce the impact of longitudinal magnetic field drift on the gyroscope scaling factor, including but not limited to the repeatability and nonlinearity of the gyroscope scaling factor.

[0009] The frequency and amplitude of the modulation magnetic field were determined after multiple experiments based on the operating state and parameters of the SERF atomic gyroscope.

[0010] The expression for the modulated magnetic field is as follows: Among them, B A ω1 is the amplitude of the modulated magnetic field, ω1 is the frequency of the modulated magnetic field, and t is time. It is the unit vector in the x-direction.

[0011] The phase demodulation expression of the lock-in amplifier is as follows:

[0012]

[0013] Where λ2 is the relaxation correlation parameter with the coupling state of alkali metal atoms and inert gas atoms, ω1 is the frequency of the applied modulation magnetic field, and γ e It is the electron gyromagnetic ratio. It is the transverse relaxation rate of alkali metal atoms, ω e This is the electron resonance frequency. The above parameters are physical constants, and only the magnetic field δB experienced by the atom is considered. z The value varies considerably with changes in environment and external parameters. θ is obtained by demodulating the phase through a lock-in amplifier when the atom senses the magnetic field δB. sWhen the signal changes, the demodulation phase θ changes accordingly. Therefore, the demodulation phase is used as the feedback quantity to realize the vertical axis atomic-sensitive vertical axis magnetic field δB. s The closed-loop control keeps it at 0.

[0014] Includes the following steps:

[0015] Step 1: Apply a sweeping sine wave with an AC amplitude of 2.5nT and a frequency of 0.01Hz to 400Hz to the x-axis coil. Demodulate the wave using a lock-in amplifier and record the peak-to-peak value at the corresponding sweeping frequency to obtain the system's electronic resonant frequency ω0.

[0016] Step 2: Stop applying the sweep frequency sine wave signal and reapply a sinusoidal modulated magnetic field with frequency ω1 = 3ω0 and amplitude 2.5nT to the x-axis coil;

[0017] Step 3: The system output signal is split into two paths. One path is connected to a low-pass filter to output a DC signal, which is used to measure the angular velocity. The other path is connected to a lock-in amplifier to perform phase demodulation based on the first harmonic of the modulated magnetic field, and the demodulated phase result is fed back to the controller. Since the atomic-sensitive longitudinal axis magnetic field δB is at this point... s If the value is equal to 0, the demodulation result at this point is defined as the initial demodulation phase, and denoted as θ0;

[0018] Step 4: Tune the proportional-integral-derivative (PID) parameters in the controller;

[0019] Step 5: The lock-in amplifier performs phase demodulation on the system output signal in real time and feeds back the real-time demodulated phase θ1 to the controller. The real-time demodulated phase θ1 is subtracted from the initial demodulated phase θ0 to generate a phase error signal θ. err ;

[0020] Step 6, the controller determines the phase error signal θ. err Generate control voltage U c And applied to the longitudinal axis compensation coil to ensure the atomic-sensitive longitudinal axis magnetic field δB s It is 0.

[0021] Step 1 includes: heating an alkali metal gas cell filled with an ensemble of K, Rb, and Ne atoms to 190°C using a non-magnetic electric heating device; locking the pump laser frequency to the D1 line of potassium atoms (770.108 nm) using a saturable absorption method; expanding the pump spot to cover the alkali metal gas cell using a beam expander; obtaining circularly polarized pump light propagating along the z-direction using a quarter-wave plate; using this light to polarize the alkali metal atoms and inert gas atoms; extracting the system output signal using a detection laser far detuned to the D1 line of Rb atoms in the x-axis direction; and after the polarization of the alkali metal atoms and inert gas atoms has stabilized, compensating the magnetic field using a magnetic field cross-modulation method via x-axis, y-axis, and z-axis coils. At this point, the gyroscope operates at gyroscope compensation point B. c0 Satisfying B c0 =-B e0 -B n0 -δB z0 B e0 For the equivalent magnetic field of electron polarization, B n0 For the nucleon polarization equivalent magnetic field, δB z0 It is triaxial remanence, at which point the atomic-sensitive vertical axis magnetic field δB s It is 0.

[0022] The SERF gyroscope system includes, in sequence, a pump laser, a concave lens, a first Glan prism, a first liquid crystal phase modulator, a second Glan prism, a convex lens, a reflector, a first λ / 2 waveplate, a third Glan prism, a first λ / 4 waveplate, an alkali metal gas cell, and a third photodetector. The third Glan prism is connected to the first liquid crystal phase modulator via the first photodetector and a first electronic control unit. The alkali metal gas cell is surrounded by, from the periphery outwards, a non-magnetic heating device, a coil, a ferrite shielding layer, a second permalloy shielding cylinder, and an outermost permalloy shielding cylinder. The coil includes a Z-axis compensation coil connected to a controller, and inner Z-axis coils, X-axis coils, and Y-axis coils connected to a signal generator. The detection light input side of the alkali metal gas cell is connected to the detection laser via a sixth Glan prism, a second λ / 2 waveplate, a fifth Glan prism, and a second liquid crystal phase modulator. The sixth Glan prism is connected to the second liquid crystal phase modulator via a second detector and a second electronic control unit. The detection light output side of the alkali metal gas cell is connected to the first input terminal of the transimpedance amplifier via a third λ / 2 waveplate, a seventh Glan prism, and a fifth photodetector. The seventh Glan prism is connected to the second input terminal of the transimpedance amplifier via a fourth photodetector. One output terminal of the transimpedance amplifier is connected to the data acquisition circuit via a low-pass filter, and the other output terminal is connected to the controller via a lock-in amplifier. The signal generator is connected to the lock-in amplifier.

[0023] The technical effects of this invention are as follows: This invention provides a closed-loop control method for the longitudinal magnetic field of a SERF atomic gyroscope. A sinusoidal modulated magnetic field, several times the electron resonance frequency, is applied in the transverse direction (x-axis). The system output signal is split into two paths: one path is input to a low-pass filter, and after filtering, it is used to output the angular velocity response signal; the other path is connected to a lock-in amplifier for phase demodulation. The phase demodulation result of the lock-in amplifier is fed back to the controller, thereby achieving closed-loop control of the longitudinal magnetic field. This method achieves both decoupled measurement of angular velocity and longitudinal magnetic field, and in-situ measurement of the longitudinal magnetic field. Since the mapping relationship with the longitudinal magnetic field is established by demodulating the phase, it is theoretically unaffected by pump laser power, detection laser power, temperature, or other parameters. Moreover, when the applied modulation frequency is higher than the electron resonance frequency, the electron spin exchange relaxation rate caused by the modulated magnetic field can be reduced, thus ensuring that the state is the same when the modulated magnetic field is applied as when there is no modulation, and significantly improving the repeatability, nonlinearity, and zero-bias stability of the gyroscope's scaling factor.

[0024] The advantages of this invention compared to the prior art are:

[0025] (1) The frequency of the modulation magnetic field applied in this invention is much greater than the electron resonance frequency, which can avoid the electron spin exchange relaxation caused by the modulation magnetic field, thereby ensuring that the scaling factor of the gyroscope for angular velocity is the same in the closed-loop state as in the open-loop state.

[0026] (2) The applied high-frequency modulated magnetic field can avoid introducing additional electron spin exchange relaxation rate, that is, it ensures that the transverse relaxation rate of alkali metal atoms is the same as that in the open-loop state, and is smaller than that in the closed-loop state based on the electron resonance peak. Therefore, theoretically, the limiting sensitivity is the same as that in the open-loop state, which is better than the limiting sensitivity in the closed-loop control state based on the electron resonance peak.

[0027] (3) Digital lock-in amplifiers have lower noise power spectral density during high-frequency demodulation. Therefore, when a high-frequency modulation magnetic field is applied, it has better resolution and correspondingly higher control accuracy compared to demodulation at the electron resonance peak. Attached Figure Description

[0028] Figure 1 A schematic diagram of the SERF atomic gyroscope system structure used to implement the SERF atomic gyroscope longitudinal axis magnetic field closed-loop control method of the present invention.

[0029] In the diagram: 1-Pump laser, 2-Concave lens, 3-First Glan prism, 4-First liquid crystal phase modulator, 5-Second Glan prism, 6-Convex lens, 7-Reflector, 8-First λ / 2 waveplate, 9-Third Glan prism, 10-First λ / 4 waveplate, 11-First electronic control unit, 12-First photodetector, 13-Second detector, 14-Second electronic control unit, 15-Detection laser, 16-Fourth Glan prism, 17-Second liquid crystal phase modulator, 18-Fifth Glan prism, 19-Second λ / 2 waveplate, 20-Sixth Glan prism, 21-Outermost permalloy shielding cylinder, 22-Second... 23-Ferrite shielding layer, 24-Non-magnetic heating device, 25-Alkali metal gas chamber, 26-Third photodetector, 27-Third λ / 2 waveplate, 28-Seventh Glan prism, 29-Fourth photodetector, 30-Data acquisition circuit (DAQ), 31-Low-pass filter (LPF), 32-Transimpedance amplifier, 33-Fifth photodetector, 34-Controller, 35-Lock-in amplifier, 36-Signal generator, 37-Z-axis compensation coil, 38-Z-axis inner coil, 39-X-axis coil, 40-Y-axis coil. Detailed Implementation

[0030] The following is in conjunction with the attached diagram ( Figure 1 The invention will be described in the following sections and examples.

[0031] Figure 1 A schematic diagram of the SERF atomic gyroscope system structure used to implement the SERF atomic gyroscope longitudinal axis magnetic field closed-loop control method of the present invention. (Reference) Figure 1As shown, a closed-loop control method for the longitudinal magnetic field of a SERF atomic gyroscope is characterized by the following steps: After achieving polarization stabilization of atoms in the alkali metal gas cell of the SERF gyroscope system, a modulation magnetic field n times the electron resonance frequency is applied to the x-axis coil to avoid the additional electron spin exchange relaxation rate introduced by the applied modulation magnetic field, thereby ensuring that the scale factor in the closed-loop control state of the gyroscope is the same as that in the open-loop state, where n is an integer greater than or equal to 2; then, the system output signal is split into two paths to achieve in-situ decoupled measurement of angular velocity and magnetic field. One path is input to a low-pass filter, and after filtering, a DC component is output, which is used to output the angular velocity response signal. The other path is connected to a lock-in amplifier for phase demodulation, and the demodulated phase value reflects the magnitude of the longitudinal magnetic field. The phase demodulation result of the lock-in amplifier is fed back to the controller, thereby realizing closed-loop control of the Z-axis magnetic field. This includes establishing a mapping relationship between the longitudinal magnetic field and the phase. The phase is theoretically unaffected by pump laser power, detection laser power, and temperature, thereby avoiding the influence of other parameter drift on the control effect. The closed-loop control of the Z-axis magnetic field can reduce the impact of longitudinal axis magnetic field drift on the gyroscope scaling factor, including but not limited to the repeatability and nonlinearity of the scaling factor. The frequency and amplitude of the modulation magnetic field were determined through multiple experiments based on the operating state and parameters of the SERF atomic gyroscope.

[0032] The expression for the modulated magnetic field is as follows: Among them, B A ω1 is the amplitude of the modulated magnetic field, ω1 is the frequency of the modulated magnetic field, and t is time. It is the unit vector in the x-direction.

[0033] The phase demodulation expression of the lock-in amplifier is as follows:

[0034]

[0035] Where λ2 is the relaxation correlation parameter with the coupling state of alkali metal atoms and inert gas atoms, ω1 is the frequency of the applied modulation magnetic field, and γ e It is the electron gyromagnetic ratio. It is the transverse relaxation rate of alkali metal atoms, ω e This is the electron resonance frequency. The above parameters are physical constants, and only the magnetic field δB experienced by the atom is considered. z The value varies considerably with changes in environment and external parameters. θ is obtained by demodulating the phase through a lock-in amplifier when the atom senses the magnetic field δB. s When the signal changes, the demodulation phase θ changes accordingly. Therefore, the demodulation phase is used as the feedback quantity to realize the vertical axis atomic-sensitive vertical axis magnetic field δB. s The closed-loop control keeps it at 0.

[0036] The process includes the following steps: Step 1, applying a sweeping sine wave with an AC amplitude of 2.5 nT and a frequency ranging from 0.01 Hz to 400 Hz to the x-axis coil, demodulating it using a lock-in amplifier, and recording the peak-to-peak value at the corresponding sweeping frequency to obtain the system's electronic resonant frequency ω0; Step 2, stopping the application of the sweeping sine wave signal, and reapplying a sinusoidal modulated magnetic field with a frequency ω1 = 3ω0 and an amplitude of 2.5 nT to the x-axis coil; Step 3, splitting the system output signal into two paths, one connected to a low-pass filter to output a DC signal for measuring angular velocity; the other connected to a lock-in amplifier, performing phase demodulation based on the first harmonic of the modulated magnetic field, and feeding the demodulated phase result back to the controller, since the atomic-sensitive longitudinal axis magnetic field δB is at this time... s The phase demodulation result is equal to 0, and is defined as the initial demodulation phase, denoted as θ0; Step 4: The proportional-integral-derivative (PID) parameters in the controller are tuned; Step 5: The lock-in amplifier performs phase demodulation on the system output signal in real time and feeds back the real-time demodulated phase θ1 to the controller. The real-time demodulated phase θ1 is subtracted from the initial demodulation phase θ0 to generate the phase error signal θ. err Step 6: The controller determines the phase error signal θ. err Generate control voltage U c And applied to the longitudinal axis compensation coil to ensure the atomic-sensitive longitudinal axis magnetic field δB s It is 0.

[0037] Step 1 includes: heating an alkali metal gas cell filled with an ensemble of K, Rb, and Ne atoms to 190°C using a non-magnetic electric heating device; locking the pump laser frequency to the D1 line of potassium atoms (770.108 nm) using a saturable absorption method; expanding the pump spot to cover the alkali metal gas cell using a beam expander; obtaining circularly polarized pump light propagating along the z-direction using a quarter-wave plate; using this light to polarize the alkali metal atoms and inert gas atoms; extracting the system output signal using a detection laser far detuned to the D1 line of Rb atoms in the x-axis direction; and after the polarization of the alkali metal atoms and inert gas atoms has stabilized, compensating the magnetic field using a magnetic field cross-modulation method via x-axis, y-axis, and z-axis coils. At this point, the gyroscope operates at gyroscope compensation point B. c0 Satisfying B c0 =-B e0 -B n0 -δB z0 B e0 For the equivalent magnetic field of electron polarization, B n0 For the nucleon polarization equivalent magnetic field, δB z0 It is triaxial remanence, at which point the atomic-sensitive vertical axis magnetic field δB s It is 0.

[0038] The SERF gyroscope system includes, in sequence, a pump laser 1, a concave lens 2, a first Glan prism 3, a first liquid crystal phase modulator 4, a second Glan prism 5, a convex lens 6, a reflector 7, a first λ / 2 waveplate 8, a third Glan prism 9, a first λ / 4 waveplate 10, an alkali metal gas chamber 25, and a third photodetector 26. The third Glan prism 9 is connected to the first liquid crystal phase modulator 4 via the first photodetector 12 and the first electronic control unit 11. The alkali metal gas chamber 25 is surrounded by, in sequence, a non-magnetic heating device 24, a coil, a ferrite shielding layer 23, a second permalloy shielding cylinder 22, and an outermost permalloy shielding cylinder 21. The coil includes a Z-axis compensation coil 37 connected to a controller 34, and an inner Z-axis coil 38, an X-axis coil 39, and a Y-axis coil 40 connected to a signal generator 36. The alkali metal gas chamber 25... The detection light input side of chamber 25 is connected to the detection laser 15 via a sixth Glan prism 20, a second λ / 2 waveplate 19, a fifth Glan prism 18, and a second liquid crystal phase modulator 17. The sixth Glan prism 20 is connected to the second liquid crystal phase modulator 17 via a second detector 13 and a second electronic control unit 14. The detection light output side of alkali metal gas chamber 25 is connected to the first input terminal of transimpedance amplifier 32 via a third λ / 2 waveplate 27, a seventh Glan prism 28, and a fifth photodetector 33. The seventh Glan prism 28 is connected to the second input terminal of transimpedance amplifier 32 via a fourth photodetector 29. One output terminal of transimpedance amplifier 32 is connected to data acquisition circuit 30 via a low-pass filter 31, and the other output terminal is connected to controller 34 via lock-in amplifier 35. The signal generator 36 is connected to lock-in amplifier 35.

[0039] The principle of this invention is as follows: After the SERF atomic spin inertial measurement system is polarized to a steady state, a modulation magnetic field is applied in the x-axis direction. Among them, B A ω1 is the amplitude of the modulated magnetic field, ω1 is the frequency of the modulated magnetic field, and t is time. It is the unit vector in the x-direction.

[0040]

[0041] Where λ2 is the relaxation correlation parameter with the coupling state of alkali metal atoms and inert gas atoms, ω1 is the frequency of the applied modulation magnetic field, and γ e It is the electron gyromagnetic ratio. It is the transverse relaxation rate of alkali metal atoms, ω e This refers to the electron resonance frequency; the above parameters are physical constants or can be considered constants. Only the magnetic field δB experienced by atoms... z It varies considerably with changes in environment and external parameters. θ is the phase demodulated by a lock-in amplifier. From the above equation, it can be seen that when the atom senses a magnetic field δB...s When the phase changes, the demodulation phase θ changes accordingly. Therefore, the demodulation phase can be used as a feedback quantity to realize the vertical axis atomic-sensitive vertical axis magnetic field δB. s The closed-loop control keeps it at 0.

[0042] A closed-loop control method for the longitudinal magnetic field of a SERF atomic spin gyroscope is proposed. After the SERF gyroscope polarization stabilizes, a sinusoidal modulated magnetic field several times the electron resonance frequency is applied to the x-axis coil. The system output signal is split into two paths. One path is input to a low-pass filter, and after filtering, it is used to output the angular velocity response signal. The other path is connected to a lock-in amplifier for phase demodulation. The phase demodulation result of the lock-in amplifier is fed back to the controller, thereby realizing closed-loop control of the Z-axis magnetic field.

[0043] The frequency of the modulated magnetic field applied in the transverse (x-axis) is higher than the electron resonant frequency, thus avoiding the additional electron spin exchange relaxation rate brought about by the applied modulated magnetic field, thereby ensuring that the scale factor of the gyroscope in the closed-loop control state is the same as that in the open-loop state.

[0044] After applying a high-frequency modulated magnetic field in the horizontal direction (x-axis), the output signal is split into two paths. One path outputs a DC component through a low-pass filter to measure the system's response to angular velocity. The other path is demodulated through a lock-in amplifier to use the demodulated phase value to reflect the magnitude of the vertical axis magnetic field. Therefore, this method achieves both decoupled measurement of angular velocity and magnetic field and in-situ measurement of the vertical axis magnetic field of the SERF atomic gyroscope.

[0045] By establishing a mapping relationship between the vertical axis magnetic field and the phase, the phase is theoretically unaffected by pump laser power, detection laser power, temperature, etc., thus avoiding the influence of other parameter drift on the control effect.

[0046] This closed-loop control method can reduce the impact of longitudinal magnetic field drift on the gyroscope scaling factor, including but not limited to the repeatability and nonlinearity of the gyroscope scaling factor.

[0047] The frequency and amplitude of the modulated magnetic field applied to the horizontal axis (x-axis) need to be measured experimentally multiple times based on the working state and parameters of the SERF atomic gyroscope.

[0048] The specific steps of this method are as follows:

[0049] Step 1: The alkali metal gas chamber 25, filled with an ensemble of K, Rb, and Ne atoms, is heated to 190°C using a non-magnetic electric heating device 24. The pump laser frequency is locked to the D1 line of potassium atoms (770.108 nm) using a saturable absorption method. A beam expander consisting of a concave lens 2 and a convex lens 6 is used to expand the pump spot to a size sufficient to cover the alkali metal gas chamber 25. Circularly polarized pump light propagating along the z-direction is obtained after passing through a first quarter-wave plate 10. This light is used to polarize the alkali metal atoms and inert gas atoms. A detection laser 15, detuned to the Rb atom D1 line, is used in the x-axis direction to extract the system output signal. After the polarization of the alkali metal atoms and inert gas atoms stabilizes, a magnetic field cross-modulation method is used to compensate the magnetic field through the x-axis coil 39, y-axis coil 40, and z-axis coil 38. At this point, the gyroscope operates at "gyroscope compensation point B". c0 ", satisfying B c0 =-B e0 -B n0 -δB z0 B e0 For the equivalent magnetic field of electron polarization, B n0 For the nucleon polarization equivalent magnetic field, δB z0 For the triaxial remanent magnetization of the shielding cylinder, the atomic sensitive magnetic field δB is at this time. s It is 0.

[0050] Step 2: Apply a sweeping sine wave with an AC amplitude of 2.5nT and a frequency of 0.01Hz to 400Hz to the x-axis coil 39. Demodulate the wave through the lock-in amplifier 35 and record the peak-to-peak value at the corresponding sweeping frequency to obtain the electronic resonant frequency ω0 of the system.

[0051] Step 3: Stop applying the sweep frequency signal and reapply a sinusoidal modulated magnetic field with frequency ω1 = 3ω0 and amplitude 2.5nT to the x-axis coil 39;

[0052] Step 4: The system output signal is split into two paths. One path is connected to a low-pass filter 31 for output, used to measure the angular velocity. The other path is connected to a lock-in amplifier 35 to perform phase demodulation based on the first harmonic of the modulated magnetic field, and the demodulated phase result is fed back to the controller. Since the atomic-sensitive longitudinal axis magnetic field δB is at this time... s If the value is 0, the demodulation result at this point is defined as the initial demodulation phase, and denoted as θ0;

[0053] Step 5: Tune the PID (Proportional-Integral-Derivative) parameters in controller 34;

[0054] Step 6: The lock-in amplifier 35 performs phase demodulation on the system output signal in real time and feeds back the real-time demodulated phase θ1 to the controller 34. The real-time demodulated phase θ1 is subtracted from the initial demodulated phase θ0 to obtain the error signal θ. err ;

[0055] Step 5, the controller 34 determines the phase error signal θ based on the phase error signal θ. err Generate control voltage U c And apply it to the Z-axis compensation coil 37 to ensure the atomic-sensitive longitudinal axis magnetic field δB s It is 0.

[0056] Contents not described in detail in this specification are prior art known to those skilled in the art. It is hereby indicated that the above description is intended to help those skilled in the art understand this invention, but does not limit the scope of protection of this invention. Any equivalent substitutions, modifications, improvements, and / or simplifications of the above descriptions that do not depart from the essential content of this invention fall within the scope of protection of this invention.

Claims

1. A closed-loop control method for the longitudinal axis magnetic field of a SERF atomic gyroscope, characterized in that, After achieving polarization stabilization of atoms in the alkali metal gas cell in the SERF gyroscope system, a modulation magnetic field n times the electron resonance frequency is applied to the x-axis coil to avoid the additional electron spin exchange relaxation rate introduced by the applied modulation magnetic field, thus ensuring that the scale factor in the closed-loop control state of the gyroscope is the same as that in the open-loop state, where n is an integer greater than or equal to 2. Then, the system output signal is split into two paths to achieve in-situ decoupled measurement of angular velocity and magnetic field. One path is input to a low-pass filter, and after filtering, a DC component is output. The DC component is used to output the angular velocity response signal. The other path is connected to a lock-in amplifier for phase demodulation. The demodulated phase value reflects the magnitude of the vertical axis magnetic field, and the phase demodulation result of the lock-in amplifier is fed back to the controller, thereby realizing closed-loop control of the Z-axis magnetic field.

2. The SERF atomic gyroscope longitudinal axis magnetic field closed-loop control method according to claim 1, characterized in that, This includes establishing a mapping relationship between the longitudinal magnetic field and the phase. The phase is theoretically unaffected by the pump laser power, the detection laser power, and the temperature, thereby avoiding the impact of other parameter drifts on the control effect.

3. The SERF atomic gyroscope longitudinal axis magnetic field closed-loop control method according to claim 1, characterized in that, The closed-loop control of the Z-axis magnetic field is used to reduce the impact of longitudinal magnetic field drift on the gyroscope scaling factor, including but not limited to the repeatability of the gyroscope scaling factor and the nonlinearity of the scaling factor.

4. The SERF atomic gyroscope longitudinal axis magnetic field closed-loop control method according to claim 1, characterized in that, The frequency and amplitude of the modulation magnetic field were determined after multiple experiments based on the operating state and parameters of the SERF atomic gyroscope.

5. The SERF atomic gyroscope longitudinal axis magnetic field closed-loop control method according to claim 1, characterized in that, The expression for the modulated magnetic field is as follows: ,in, It modulates the amplitude of the magnetic field. It refers to the frequency of the modulated magnetic field, where t is time. It is the unit vector in the x-direction.

6. The SERF atomic gyroscope longitudinal axis magnetic field closed-loop control method according to claim 1, characterized in that, The phase demodulation expression of the lock-in amplifier is as follows: ， in It is a relaxation-related parameter associated with the coupling state of alkali metal atoms and inert gas atoms. It is the frequency of the applied modulating magnetic field. It is the electron gyromagnetic ratio. It is the transverse relaxation rate of alkali metal atoms. It is the electron resonance frequency. The above parameters are physical constants. Only the atomic-sensitive longitudinal axis magnetic field... It varies considerably with changes in environment and external parameters. It demodulates the phase through a lock-in amplifier, when... When changing, As this changes, the demodulated phase is used as the feedback quantity to realize the atomically sensitive vertical axis magnetic field. The closed-loop control keeps it at 0.

7. The SERF atomic gyroscope longitudinal axis magnetic field closed-loop control method according to claim 1, characterized in that, Includes the following steps: Step 1: Apply a sweeping sine wave with an AC amplitude of 2.5nT and a frequency ranging from 0.01Hz to 400Hz to the x-axis coil. Demodulate the wave using a lock-in amplifier and record the peak-to-peak value at the corresponding sweeping frequency to obtain the system's electronic resonant frequency. ; Step 2: Stop applying the sweep sine wave signal and reapply the frequency to the x-axis coil. A sinusoidal modulated magnetic field with an amplitude of 2.5 nT; Step 3: The system output signal is split into two paths. One path is connected to a low-pass filter to output a DC signal, which is used to measure the angular velocity. The other path is connected to a lock-in amplifier to perform phase demodulation based on the first harmonic of the modulated magnetic field, and the demodulated phase result is fed back to the controller. Since the atomic-sensitive longitudinal axis magnetic field is at this point... When the value equals 0, the phase demodulation result at this point is defined as the initial demodulation phase, and denoted as . ; Step 4: Tune the proportional-integral-derivative (PID) parameters in the controller; Step 5: The lock-in amplifier performs phase demodulation on the system output signal in real time, and transmits the real-time demodulated phase... Feedback is sent to the controller for real-time phase demodulation. With the initial demodulation phase Subtraction generates a phase error signal. ; Step 6, the controller determines the phase error signal. Generate control voltage And applied to the longitudinal axis compensation coil to ensure the atomic-sensitive longitudinal axis magnetic field. It is 0.

8. The SERF atomic gyroscope longitudinal axis magnetic field closed-loop control method according to claim 7, characterized in that, Step 1 includes: heating an alkali metal gas cell filled with an ensemble of K, Rb, and Ne atoms to 190°C using a non-magnetic electric heating device; locking the pump laser frequency to the D1 line of potassium atoms (770.108 nm) using a saturable absorption method; expanding the pump spot to cover the alkali metal gas cell using a beam expander; obtaining circularly polarized pump light propagating along the z-direction using a quarter-wave plate; using this light to polarize the alkali metal atoms and inert gas atoms; extracting the system output signal using a detection laser far detuned to the D1 line of Rb atoms in the x-axis direction; and after the polarization of the alkali metal atoms and inert gas atoms has stabilized, compensating the magnetic field using a magnetic field cross-modulation method via x-axis, y-axis, and z-axis coils, at which point the gyroscope operates at the gyroscope compensation point. ,satisfy ,in This is the equivalent magnetic field for electron polarization. The equivalent magnetic field for nucleon polarization, It is a triaxial remanence, at which point the atom is sensitive to the vertical axis magnetic field. It is 0.

9. The SERF atomic gyroscope longitudinal axis magnetic field closed-loop control method according to claim 1, characterized in that, The SERF gyroscope system includes, in sequence, a pump laser, a concave lens, a first Glan prism, a first liquid crystal phase modulator, a second Glan prism, a convex lens, a reflector, and a first... Wave plate, third Glan prism, first The system comprises a waveplate, an alkali metal gas chamber, and a third photodetector. The third Glan prism is connected to the first liquid crystal phase modulator via the first photodetector and the first electronic control unit. The alkali metal gas chamber is surrounded by a non-magnetic heating device, a coil, a ferrite shielding layer, a second permalloy shielding cylinder, and an outermost permalloy shielding cylinder. The coil includes a Z-axis compensation coil connected to the controller, and inner Z-axis coils, X-axis coils, and Y-axis coils connected to the signal generator. The detection light input side of the alkali metal gas chamber is connected to a sixth Glan prism, the second… A waveplate, a fifth Glan prism, and a second liquid crystal phase modulator are connected to the detection laser. The sixth Glan prism is connected to the second liquid crystal phase modulator via a second detector and a second electronic control unit. The detection light output side of the alkali metal gas cell is connected to a third... A waveplate, a seventh Glan prism, and a fifth photodetector are connected to the first input terminal of a transimpedance amplifier. The seventh Glan prism is connected to the second input terminal of the transimpedance amplifier via a fourth photodetector. One output terminal of the transimpedance amplifier is connected to a data acquisition circuit via a low-pass filter, and the other output terminal is connected to a controller via a lock-in amplifier. The signal generator is connected to the lock-in amplifier.

Images