Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

IEEE Data Driven Control and Learning Systems Conference (DDCLS)

A Data-driven Physical Mechanism Modeling Method for the Spin-Exchange Relaxation-Free Comagnetometer

Feng Li, Zhuo Wang, Min Zhang, Ruigang Wang, Bodong Qin, Yanchao Chai

Published May 12, 2023• 0 citations
Publisher

Abstract

The Spin-Exchange Relaxation-Free Comagnetometer (SERFCM) is a new quantum instrument with ultra-high accuracy. Normally, the atomic ensembles of SERFCM operate in an open-loop state, which is not conducive to long-term high-precision measurements. In order to realize closed-loop control of its atomic polarization state, it is necessary to model and analyze the dynamic characteristics of the SERFCM system. In this paper, a Data-driven physical mechanism (DDPM) modeling method is proposed to realize the modeling of the SERFCM, a multi-input multi-output system. First, the state space equations of the SERFCM are established based on the Bloch equation, which are transformed into a discrete transfer function matrix. Then, based on the criterion of least variance in estimation, we realize the modeling of the discrete transfer function matrix using the excitation input data, the measured output data, and the estimated output data. Finally, the simulation results of modeling under different longitudinal magnetic fields confirm the validity of the proposed method. This work enables the online modeling of SERFCM system and facilitates the analysis of the effects of various parameters on the dynamic characteristics.

DOI
10.1109/ddcls58216.2023.10167072

Related papers

Atomic Spin Polarization Controllability Analysis: A Novel Controllability Determination Method for Spin-Exchange Relaxation-Free Co-Magnetometers

Zhuo Wang, Sixun Liu, Ruigang Wang +4 more

This paper investigates the atomic spin polarization controllability of spin-exchange relaxation-free co-magnetometers (SERFCMs). This is the first work in the field of controllability analysis for the atomic spin ensembles systems, whose dynamic behaviors of spin polarization are described by the Bloch equations. Based on the Bloch equations, a state-space model of the atomic spin polarization for SERFCM is first established, which belongs to a particular class of nonlinear systems. For this class of nonlinear systems, a novel determination method for the global state controllability is proposed and proved. Then, this method is implemented in the process of controllability analysis on the atomic spin polarization of an actual SERFCM. Moreover, a theoretically feasible and reasonable solution of the control input is proposed under some physical constraints, with whose limitation of realistic conditions, the controller design can be accomplished more practically and more exactly. Finally, the simulation results demonstrate the feasibility and validation of the proposed controllability determination method.

IEEE/CAA Journal of Automatica Sinica•Dec 10, 2021•33 citations

Analysis and Suppression of the Pump Light Intensity Error in the K–Rb– 21 Ne Atomic Spin Comagnetometer

Jiale Quan, Ye Liu, Longyan Ma +5 more

As a novel quantum sensor utilizing atomic spin coupling mechanisms, spin-exchange relaxation-free (SERF) comagnetometers demonstrate promising applications in precision metrology such as magnetic field and rotation sensing. However, the ultra-high precision measurement capability exhibits vulnerability to multiple disturbance factors, with pump light intensity fluctuations emerging as a critical instability source that compromises operational reliability during prolonged measurements. In this paper, a model-driven paradigm for pump light intensity error suppression in SERF comagnetometers is proposed. The nonlinear dynamics are rigorously linearized via Taylor expansion within their valid range to construct a state-space model, ultimately deriving the pump light intensity frequency response model. The proposed methodology undergoes experimental validation on a K-Rb-21Ne comagnetometry platform, confirming the predictive capability for stability degradation mechanisms. Through theoretical analysis and experimental verification, this study reveals that magneto-optical non-orthogonality, which introduces a differential component in the light intensity transfer function, creates an approximate proportional relationship between the output signal and pump light intensity within the operating bandwidth. Light intensity noise is decomposed, and light intensity errors are quantitatively evaluated via the frequency response model. To address this critical issue affecting inertial measurement accuracy, a closed-loop control-based suppression method with optimized feedback gain is proposed. By dynamically adjusting the closed-loop feedback gain through real-time intensity monitoring, the intensity instability associated with optical pumping is suppressed from 0.56% to 0.14%, concomitant with a 2.8-fold improvement in the long-term stability of the SERF comagnetometer.

IEEE Transactions on Instrumentation and Measurement•Jan 1, 2025•2 citations

Dynamically Polarized SERF Atomic Comagnetometer

Xiaofei Huang, Wei, Kai, Rui Yang +4 more

Atomic spin sensors are essential for beyond-the-standard-model exploration, biomagnetic measurement, and quantum navigation. While the traditional DC mode spin-exchange relaxation-free (SERF) comagnetometer achieves ultrahigh sensitivity, further improvements require suppressing technical noise and surpassing standard quantum limit. In this work, we develop a K-Rb-$^{21}$Ne SERF atomic comagnetometer that dynamically polarizes the electron and nuclear spins, shielding signals from direct interference by pump light. We establish a three-phase evolutionary model for hybrid spin ensemble dynamics, yielding a complete analytical solution, and analyze the responses to various spin perturbations. Additionally, we achieve an averaged 38.5 $\%$ suppression of the polarization noise and identify the key factors that limit sensitivity improvements. The dynamically polarized comagnetometer exhibits effective suppression of technical noise and holds the potential to overcome quantum noise limit, while offering promising applications in exploring new physics and precise magnetic field measurements.

arXiv:2505.18462•May 24, 2025•0 citations

Automatic magnetic compensation system for SERF atomic comagnetometer

Zhuo Wang, Luxv Zhai, Jiong Huang

To realize the SERF (spin exchange relaxation free) state of the atomic comagnetometer for inertial measurement, this paper proposes an efficient automatic magnetic compensation system based on the analysis of the dynamic model of the SERF atomic comagnetometer, such that the essential and necessary condition of low environmental magnetic field is satisfied. This system first makes a fast but rough compensation to reduce the impact of environmental magnetic field to a very low level, then accurately but slowly compensates the residual magnetic field. The rough compensation is made under low power detection light to achieve high efficiency, while the accurate one is made under high power detection light to achieve high accuracy, which are both automatically performed. This is definitely an advantage over the most existing magnetic compensation systems for the SERF atomic comagnetometer, which rely only on manual adjustments with low accuracy and poor repeatability. The simulation illustrates the good efficiency, accuracy and repeatability of the proposed automatic magnetic compensation system.

Youth Academic Annual Conference of Chinese Association of Automation (YAC)•Oct 16, 2020•0 citations

Reachability Analysis of The Atomic Spin Polarization for Spin-Exchange Relaxation-Free Atomic Magnetometers

Ruigang Wang, Zhuo Wang, Chong Ma +4 more

This paper is the first work to innovatively analyze and determine the reachability of atomic spin polarization of Spin-Exchange Relaxation-Free Atomic Magnetometers (SERFAMs), whose dynamic behaviors described by the Bloch equations. Then, it is transformed into a cybernetic model to analyze the reachability. The results show that the degree of freedom of state transition of atomic spin polarization is not consistent with the input dimension. Although this is contrary to the physical intuition, it corresponds to the point of cybernetics. The simulation results verify this conclusion. This work is an interesting case of applying cybernetics to the study of physics, which broadens the way of thinking for the performance analysis of atomic spin polarization system.

International Conference on Control and Robotics (ICCR)•Dec 2, 2022•1 citation

Dynamically polarized atomic comagnetometer

Xiaofei Huang, Kai Wei, Yang Rui +4 more

Atomic spin sensors are essential for beyond-the-standard-model exploration, biomagnetic measurement, and quantum navigation. While the traditional DC mode spin-exchange relaxation-free (SERF) comagnetometer achieves ultrahigh sensitivity, further improvements require suppressing technical noise and surpassing standard quantum limit. In this work, we develop a K-Rb- 21 Ne SERF atomic comagnetometer that dynamically polarizes the electron and nuclear spins, shielding signals from direct interference by pump light. We establish a three-phase evolutionary model for hybrid spin ensemble dynamics, yielding a complete analytical solution, and analyze the responses to various spin perturbations. Additionally, we achieve an average 39% suppression of the polarization noise and identify the key factors that limit sensitivity improvements. The dynamically polarized comagnetometer exhibits effective suppression of technical noise and holds the potential to overcome quantum noise limit while offering promising applications in exploring new physics and precise magnetic field measurements.

Cell Reports Physical Science•Nov 1, 2025•1 citation
The Spin-Exchange Relaxation-Free Comagnetometer (SERFCM) is a new quantum instrument with ultra-high accuracy. Normally, the atomic ensembles of SERFCM operate in an open-loop state, which is not conducive to long-term high-precision measurements. In order to realize closed-loop control of its atomic polarization state, it is necessary to model and analyze the dynamic characteristics of the SERFCM system. In this paper, a Data-driven physical mechanism (DDPM) modeling method is proposed to realize the modeling of the SERFCM, a multi-input multi-output system. First, the state space equations of the SERFCM are established based on the Bloch equation, which are transformed into a discrete transfer function matrix. Then, based on the criterion of least variance in estimation, we realize the modeling of the discrete transfer function matrix using the excitation input data, the measured output data, and the estimated output data. Finally, the simulation results of modeling under different longitudinal magnetic fields confirm the validity of the proposed method. This work enables the online modeling of SERFCM system and facilitates the analysis of the effects of various parameters on the dynamic characteristics.
This paper investigates the atomic spin polarization controllability of spin-exchange relaxation-free co-magnetometers (SERFCMs). This is the first work in the field of controllability analysis for the atomic spin ensembles systems, whose dynamic behaviors of spin polarization are described by the Bloch equations. Based on the Bloch equations, a state-space model of the atomic spin polarization for SERFCM is first established, which belongs to a particular class of nonlinear systems. For this class of nonlinear systems, a novel determination method for the global state controllability is proposed and proved. Then, this method is implemented in the process of controllability analysis on the atomic spin polarization of an actual SERFCM. Moreover, a theoretically feasible and reasonable solution of the control input is proposed under some physical constraints, with whose limitation of realistic conditions, the controller design can be accomplished more practically and more exactly. Finally, the simulation results demonstrate the feasibility and validation of the proposed controllability determination method.
As a novel quantum sensor utilizing atomic spin coupling mechanisms, spin-exchange relaxation-free (SERF) comagnetometers demonstrate promising applications in precision metrology such as magnetic field and rotation sensing. However, the ultra-high precision measurement capability exhibits vulnerability to multiple disturbance factors, with pump light intensity fluctuations emerging as a critical instability source that compromises operational reliability during prolonged measurements. In this paper, a model-driven paradigm for pump light intensity error suppression in SERF comagnetometers is proposed. The nonlinear dynamics are rigorously linearized via Taylor expansion within their valid range to construct a state-space model, ultimately deriving the pump light intensity frequency response model. The proposed methodology undergoes experimental validation on a K-Rb-21Ne comagnetometry platform, confirming the predictive capability for stability degradation mechanisms. Through theoretical analysis and experimental verification, this study reveals that magneto-optical non-orthogonality, which introduces a differential component in the light intensity transfer function, creates an approximate proportional relationship between the output signal and pump light intensity within the operating bandwidth. Light intensity noise is decomposed, and light intensity errors are quantitatively evaluated via the frequency response model. To address this critical issue affecting inertial measurement accuracy, a closed-loop control-based suppression method with optimized feedback gain is proposed. By dynamically adjusting the closed-loop feedback gain through real-time intensity monitoring, the intensity instability associated with optical pumping is suppressed from 0.56% to 0.14%, concomitant with a 2.8-fold improvement in the long-term stability of the SERF comagnetometer.
Atomic spin sensors are essential for beyond-the-standard-model exploration, biomagnetic measurement, and quantum navigation. While the traditional DC mode spin-exchange relaxation-free (SERF) comagnetometer achieves ultrahigh sensitivity, further improvements require suppressing technical noise and surpassing standard quantum limit. In this work, we develop a K-Rb-21^{21}21Ne SERF atomic comagnetometer that dynamically polarizes the electron and nuclear spins, shielding signals from direct interference by pump light. We establish a three-phase evolutionary model for hybrid spin ensemble dynamics, yielding a complete analytical solution, and analyze the responses to various spin perturbations. Additionally, we achieve an averaged 38.5 %\%% suppression of the polarization noise and identify the key factors that limit sensitivity improvements. The dynamically polarized comagnetometer exhibits effective suppression of technical noise and holds the potential to overcome quantum noise limit, while offering promising applications in exploring new physics and precise magnetic field measurements.
To realize the SERF (spin exchange relaxation free) state of the atomic comagnetometer for inertial measurement, this paper proposes an efficient automatic magnetic compensation system based on the analysis of the dynamic model of the SERF atomic comagnetometer, such that the essential and necessary condition of low environmental magnetic field is satisfied. This system first makes a fast but rough compensation to reduce the impact of environmental magnetic field to a very low level, then accurately but slowly compensates the residual magnetic field. The rough compensation is made under low power detection light to achieve high efficiency, while the accurate one is made under high power detection light to achieve high accuracy, which are both automatically performed. This is definitely an advantage over the most existing magnetic compensation systems for the SERF atomic comagnetometer, which rely only on manual adjustments with low accuracy and poor repeatability. The simulation illustrates the good efficiency, accuracy and repeatability of the proposed automatic magnetic compensation system.
This paper is the first work to innovatively analyze and determine the reachability of atomic spin polarization of Spin-Exchange Relaxation-Free Atomic Magnetometers (SERFAMs), whose dynamic behaviors described by the Bloch equations. Then, it is transformed into a cybernetic model to analyze the reachability. The results show that the degree of freedom of state transition of atomic spin polarization is not consistent with the input dimension. Although this is contrary to the physical intuition, it corresponds to the point of cybernetics. The simulation results verify this conclusion. This work is an interesting case of applying cybernetics to the study of physics, which broadens the way of thinking for the performance analysis of atomic spin polarization system.
Atomic spin sensors are essential for beyond-the-standard-model exploration, biomagnetic measurement, and quantum navigation. While the traditional DC mode spin-exchange relaxation-free (SERF) comagnetometer achieves ultrahigh sensitivity, further improvements require suppressing technical noise and surpassing standard quantum limit. In this work, we develop a K-Rb- 21 Ne SERF atomic comagnetometer that dynamically polarizes the electron and nuclear spins, shielding signals from direct interference by pump light. We establish a three-phase evolutionary model for hybrid spin ensemble dynamics, yielding a complete analytical solution, and analyze the responses to various spin perturbations. Additionally, we achieve an average 39% suppression of the polarization noise and identify the key factors that limit sensitivity improvements. The dynamically polarized comagnetometer exhibits effective suppression of technical noise and holds the potential to overcome quantum noise limit while offering promising applications in exploring new physics and precise magnetic field measurements.