arxiv: 2603.25327 · v1 · submitted 2026-03-26 · 🌌 astro-ph.IM

Recognition: no theorem link

Calibration of key parameters during the in-orbit phase for the Taiji-2 gravitational reference sensor

Haoyue Zhang , Chang Liu , Xiaotong Wei , Peng Xu , Li-e Qiang , Ziren Luo , Ye Dong

Authors on Pith no claims yet

Pith reviewed 2026-05-15 00:47 UTC · model grok-4.3

classification 🌌 astro-ph.IM

keywords in-orbit calibrationgravitational reference sensorKalman filterscale factorcenter of mass offsettorque maneuverTaiji missiongravitational wave detection

0 comments

The pith

Periodic torques and Kalman filtering calibrate Taiji-2 GRS scale factors below 0.2% while fixing center-of-mass offsets to 100 micrometers.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops an in-orbit method to keep the gravitational reference sensors on Taiji-2 at the precision required for gravitational wave detection. Scale factors and center-of-mass offsets between test masses and the spacecraft drift over time from propellant use, temperature changes, and electronics aging. The solution applies periodic torque signals to generate controlled angular accelerations, then feeds star tracker and sensor readouts into a Kalman filter that separates the coupled effects and estimates both parameters at once. The resulting errors stay below 0.2% for scale factors and within 100 micrometers for offsets, meeting mission needs across different spacecraft setups. This keeps the sensors capable of reaching the target acceleration sensitivity of 3 times 10 to the minus 15 meters per second squared per square root Hertz.

Core claim

By applying periodic torque signals to induce controlled spacecraft angular accelerations, the method leverages star tracker and GRS readouts to disentangle coupled disturbances and achieves dual-parameter calibration with scale factors errors below 0.2% and c.m. offsets residuals within 100 μm, satisfying the Taiji-2 calibration requirements.

What carries the argument

Kalman filter that jointly estimates GRS scale factors and center-of-mass offsets from periodic torque-induced angular accelerations observed by star trackers and gravitational reference sensors.

If this is right

The calibrated sensors maintain the 3×10^{-15} m s^{-2} Hz^{-1/2} sensitivity needed for Taiji-2 gravitational wave observations.
The dual-parameter estimates ensure the mission's scientific objectives remain feasible despite ongoing drifts.
The approach remains effective across varied satellite mass and configuration changes.
The same framework supplies a calibration path for later missions that demand sub-micrometer center-of-mass stability.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Torque-maneuver calibration could transfer to other space interferometers that must track drifting sensor parameters over long missions.
Filter residuals from unmodeled effects such as sloshing could serve as diagnostics to refine the model further.
Onboard automation of the periodic torque sequence might allow self-calibration without frequent ground intervention.

Load-bearing premise

The Kalman filter model captures every relevant dynamic and the applied torques produce accelerations known to high accuracy with no unmodeled disturbances from propellant sloshing, thermal gradients, or aging electronics.

What would settle it

Flight data showing scale factor errors above 0.2% or center-of-mass residuals above 100 micrometers after the torque maneuvers and filter processing would disprove the claimed calibration precision.

Figures

Figures reproduced from arXiv: 2603.25327 by Chang Liu, Haoyue Zhang, Li-e Qiang, Peng Xu, Xiaotong Wei, Ye Dong, Ziren Luo.

**Figure 2.** Figure 2: FIG. 2: Layout of the electrodes of the GRS. [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗

**Figure 3.** Figure 3: FIG. 3: SC c.m. location and TM distribution (Type-1). [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗

**Figure 4.** Figure 4: FIG. 4: SC c.m. location and TM distribution (Type-2). [PITH_FULL_IMAGE:figures/full_fig_p003_4.png] view at source ↗

**Figure 5.** Figure 5: FIG. 5: Flowchart of Taiji-2 calibration scheme. [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗

**Figure 6.** Figure 6: FIG. 6: The simulated ST readout of maneuvering [PITH_FULL_IMAGE:figures/full_fig_p005_6.png] view at source ↗

**Figure 7.** Figure 7: FIG. 7: The simulated ST readout of maneuvering [PITH_FULL_IMAGE:figures/full_fig_p006_7.png] view at source ↗

**Figure 8.** Figure 8: FIG. 8: The fit results of GRS1 and simulated ST [PITH_FULL_IMAGE:figures/full_fig_p008_8.png] view at source ↗

**Figure 10.** Figure 10: FIG. 10: The fit results of simulated GRS1 readouts for [PITH_FULL_IMAGE:figures/full_fig_p008_10.png] view at source ↗

**Figure 9.** Figure 9: FIG. 9: The calibration results of scale factors of GRS1 [PITH_FULL_IMAGE:figures/full_fig_p008_9.png] view at source ↗

**Figure 11.** Figure 11: FIG. 11: The c.m. offsets calibration results of GRS1 [PITH_FULL_IMAGE:figures/full_fig_p008_11.png] view at source ↗

read the original abstract

The Taiji mission, a pioneering Chinese space-borne gravitational wave observatory, requires ultra-precise calibration of its gravitational reference sensors (GRSs) to achieve its targeted sensitivity of $3\times10^{-15} \mathrm{\ m\ s^{-2}\ Hz^{-1/2}}$. Maintaining this precision is challenged by time-varying scale factors drifts and dynamic center-of-mass (c.m.) offsets between the test masses (TMs) and spacecraft, driven by factors such as propellant consumption, thermal effects and aging electronics. This paper develops an advanced in-orbit calibration framework that simultaneously estimates the GRS scale factors and c.m. offsets between TMs and spacecraft through a combination of spacecraft maneuvers and Kalman filter. By applying periodic torque signals to induce controlled spacecraft angular accelerations, we leverage star tracker and GRS readouts to disentangle coupled disturbances and achieve dual-parameter calibration with unprecedented precision, with scale factors errors below 0.2\% and c.m. offsets residuals within 100 $\mathrm{\mu}$m, satisfies the Taiji-2 calibration requirements. This method is robust across different satellite configurations. The results not only ensure the feasibility of Taiji-2's scientific objectives but also establish a scalable calibration paradigm for future missions such as Taiji-3, where sub-micrometer c.m. stability and ultra-low noise gravitational reference will be essential.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Simulation of torque maneuvers plus Kalman filter gives a workable in-orbit GRS calibration plan for Taiji-2, but the quoted precision rests on untested model completeness.

read the letter

The main thing to know is that this paper runs a simulation of using periodic torque commands to induce spacecraft angular accelerations, then feeds star-tracker and GRS readouts into a Kalman filter to estimate scale-factor drifts and center-of-mass offsets at the same time. They report simulation residuals below 0.2 % for the scale factors and 100 μm for the offsets, which they say meets Taiji-2 requirements and holds across different satellite layouts. That is the concrete engineering piece they contribute. The method is a direct application of standard estimation tools to the specific problem of time-varying GRS parameters driven by propellant use, thermal shifts, and electronics aging. It is useful because it avoids separate calibration steps and ties the maneuvers straight to the mission noise budget. The soft spot is exactly the one the stress-test note flags: everything depends on the filter model containing every term that couples into the measurements and on the commanded torques being known to high accuracy. The abstract lists propellant sloshing, thermal gradients, and aging electronics as sources of varying offsets, yet the simulation only includes the disturbances the authors chose to inject. Any real term left out of the dynamics matrix will bias the separation of the two parameters. There are no error budgets, no sensitivity runs with omitted disturbances, and no comparison against independent validation data shown in the material I have. The citation pattern is standard and does not hide prior work. This paper is for the Taiji team and for groups building similar space-based gravitational-wave sensors who need a concrete maneuver-and-filter recipe. A reader working on spacecraft dynamics or GRS calibration will get practical value from the maneuver design even if they later modify the filter. It is coherent on its own terms and shows clear thinking about the mission constraints, so it deserves a serious referee to check the dynamics matrix, the process-noise tuning, and how much margin exists when unmodeled effects appear. I would send it to peer review.

Referee Report

2 major / 2 minor

Summary. The paper presents an in-orbit calibration framework for the Taiji-2 gravitational reference sensors (GRS). It applies periodic torque signals to induce controlled spacecraft angular accelerations and uses a Kalman filter that fuses star-tracker and GRS readouts to simultaneously estimate GRS scale factors and center-of-mass (c.m.) offsets between test masses and spacecraft. Simulations are used to demonstrate that the method achieves scale-factor errors below 0.2 % and c.m. offset residuals within 100 μm while remaining robust across different satellite configurations, thereby satisfying Taiji-2 requirements.

Significance. If the simulation results translate to flight, the approach would provide a practical, scalable solution for maintaining the sub-femto-g sensitivity required by Taiji-class gravitational-wave observatories. The dual-parameter estimation via controlled maneuvers is a clear strength, and the reported residuals meet the mission’s stated calibration targets. However, the significance is tempered by the absence of a full error budget and by the reliance on an unvalidated assumption that all relevant disturbances are captured in the filter model.

major comments (2)

The headline performance (scale-factor errors <0.2 % and c.m. residuals <100 μm) rests on the Kalman filter’s ability to separate the two parameter sets. The manuscript does not demonstrate that the filter state vector includes all torque disturbances (propellant sloshing, thermal gradients, aging electronics) listed in the abstract as sources of time-varying c.m. offsets. If any such term is omitted from the dynamics matrix, the least-squares separation will be biased; the simulation residuals therefore reflect only the disturbances that were deliberately injected.
No quantitative error budget or torque-command accuracy specification is provided. The method assumes that the applied periodic torques produce accelerations whose magnitude and direction are known to high accuracy; without an analysis of actuator noise, misalignment, or propellant-induced torque errors, it is unclear whether the quoted residuals remain achievable when the commanded torques themselves carry realistic uncertainty.

minor comments (2)

The abstract states that the method “satisfies the Taiji-2 calibration requirements,” but the manuscript should explicitly compare the achieved residuals against the numerical requirements given in the Taiji-2 mission documents (e.g., maximum allowable scale-factor drift and c.m. stability).
Notation for the Kalman-filter process and measurement noise covariances is introduced without a clear table or appendix listing the numerical values used in the simulations; reproducibility would be improved by providing these values.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed comments. We have prepared point-by-point responses below and will revise the manuscript accordingly to improve clarity and completeness.

read point-by-point responses

Referee: The headline performance (scale-factor errors <0.2 % and c.m. residuals <100 μm) rests on the Kalman filter’s ability to separate the two parameter sets. The manuscript does not demonstrate that the filter state vector includes all torque disturbances (propellant sloshing, thermal gradients, aging electronics) listed in the abstract as sources of time-varying c.m. offsets. If any such term is omitted from the dynamics matrix, the least-squares separation will be biased; the simulation residuals therefore reflect only the disturbances that were deliberately injected.

Authors: The Kalman filter state vector incorporates the dominant torque disturbances relevant to Taiji-2, specifically those arising from propellant consumption, thermal gradients, and electronics aging that produce the primary time-varying c.m. offsets. The simulations inject representative realizations of these modeled terms to validate parameter separation. We acknowledge that the manuscript would benefit from greater explicitness and will revise Section 3 to list the exact state-vector components, justify the modeled disturbances against mission specifications, and add a brief analysis showing that higher-order unmodeled effects remain below the 0.2 % / 100 μm thresholds under the assumed noise levels. revision: partial
Referee: No quantitative error budget or torque-command accuracy specification is provided. The method assumes that the applied periodic torques produce accelerations whose magnitude and direction are known to high accuracy; without an analysis of actuator noise, misalignment, or propellant-induced torque errors, it is unclear whether the quoted residuals remain achievable when the commanded torques themselves carry realistic uncertainty.

Authors: We agree that a quantitative error budget is required to substantiate the torque-command assumptions. In the revised manuscript we will add a dedicated error-budget subsection that propagates actuator noise, misalignment angles, and propellant-induced torque uncertainties through the Kalman filter. Additional Monte-Carlo simulations will be presented demonstrating that the resulting scale-factor and c.m.-offset residuals remain within the stated limits for realistic torque-command errors consistent with Taiji-2 actuator specifications. revision: yes

Circularity Check

0 steps flagged

No circularity: calibration derives from external star-tracker data and torque commands via standard Kalman estimation

full rationale

The paper's method applies known periodic torques to induce accelerations, then uses independent star-tracker and GRS readouts in a Kalman filter to jointly estimate scale factors and c.m. offsets. No equation reduces the claimed outputs (scale-factor error <0.2%, c.m. residual <100 μm) to the inputs by construction, nor renames a fitted parameter as a prediction. No self-citation is invoked as a uniqueness theorem or load-bearing premise for the central result. The simulation results reflect injected disturbances within the modeled dynamics; the derivation chain remains self-contained against external measurements and does not collapse to self-definition or ansatz smuggling.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The framework rests on standard spacecraft attitude dynamics and Kalman-filter assumptions; no new physical entities are postulated.

free parameters (1)

Kalman filter process and measurement noise covariances
These matrices must be tuned to the expected disturbance spectrum and are not derived from first principles.

axioms (1)

domain assumption Spacecraft angular accelerations induced by commanded torques are known to the accuracy required by the filter
Invoked when the method treats the applied torques as perfectly known inputs.

pith-pipeline@v0.9.0 · 5563 in / 1209 out tokens · 36669 ms · 2026-05-15T00:47:34.860213+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

30 extracted references · 30 canonical work pages · 1 internal anchor

[1]

As shown in Tab

1 × 10− 11( 3 mHz f ) m s − 2 Hz− 1/ 2 into the GRS [ 17]. As shown in Tab. II, the linear acceleration read by the GRS is linearly related to its voltage output, with the slope deﬁned as the scale factors. Consequently, voltage ﬂuc- tuations equivalent to SRP ﬂuctuations denote SV = SδSRP Greal , (1) where G real is the true scale factors of the GRS. Whe...

work page
[2]

The TMRF is deﬁned by the set {Omi, xi, yi, zi}, i = {1, 2}. The complete expression describes the acceleration of the i-th TM during ﬂight, as observed in the CRF when the system operates in accelerometer mode: ¨⃗ rmi = m − 1 ( ⃗FEi + ⃗Dmi + ⃗FSi ) , (4) here m denotes the mass of the TM, ⃗FE is electrostatic suspension force, and ⃗Dm represents the dist...

work page
[3]

(12) From Eq

Thus acceleration noise induced by attitude of SC could be simpliﬁed as: ⃗ acomi =Cmi, C( ˙ω SC × ⃗di + ω SC × (ω SC × ⃗di)). (12) From Eq. 12, it can be seen that by carefully calibrat- ing ⃗di, the ⃗ acomi signal can be accurately identiﬁed and eliminated from the GRS readouts. In Eq. 12, the sec- ond term is a second-order small quantity, and thus the ...

work page
[4]

TABLE III: The payload noise considered during the calibration experiment

1 × 10− 11( 3 mHz f ) m s − 2 Hz− 1/ 2. TABLE III: The payload noise considered during the calibration experiment. LFST and LFSR respectively denote the residual translational and rotational acceleration noises from LFS control. Noise sources Nominal Value ST readout noise ≤ 0. 2′′ Thruster noise ≤ 0. 1 µ N Hz − 1/ 2 LFSTx ≤ 2. 4 × 10− 12 m s − 2 Hz− 1/ 2...

work page
[5]

Here, SIM de- notes the simulated readouts, and KF represents the re- sults obtained using the Kalman ﬁlter. In Fig. 8, the blue line represents the simulation readouts from the ST. The orange line indicates the ﬁtted results obtained by com- bining the GRS readouts with the Kalman ﬁlter. The Kalman ﬁlter results of G angle are shown in Fig. 9, the blue l...

work page
[6]

B. P. Abbott, R. Abbott, T. D. Abbott, M. R. Aber- nathy, F. Acernese, K. Ackley, C. Adams, T. Adams, P. Addesso, R. X. Adhikari, V. B. Adya, et al. (LIGO Scientiﬁc Collaboration and Virgo Collaboration), Ob- servation of gravitational waves from a binary black hole merger, Phys. Rev. Lett. 116, 061102 (2016)

work page 2016
[7]

Laser Interferometer Space Antenna

P. Amaro-Seoane, H. Audley, S. Babak, J. Baker, E. Ba- rausse, P. Bender, E. Berti, P. Binetruy, M. Born, D. Bor- toluzzi, J. Camp, et al. , Laser interferometer space an- tenna, arXiv 1702.00786 (2017)

work page internal anchor Pith review Pith/arXiv arXiv 2017
[8]

Hu and Y.-L

W.-R. Hu and Y.-L. Wu, The Taiji Program in Space for gravitational wave physics and the nature of gravity, Natl. Sci. Rev. 4, 685 (2017)

work page 2017
[9]

Luo, L.-S

J. Luo, L.-S. Chen, H.-Z. Duan, Y.-G. Gong, S. Hu, J. Ji, Q. Liu, J. Mei, V. Milyukov, M. Sazhin, C.-G. Shao, V. T. Toth, H.-B. Tu, Y. Wang, Y. Wang, H.- C. Yeh, M.-S. Zhan, Y. Zhang, V. Zharov, and Z.-B. Zhou, Tianqin: a space-borne gravitational wave detec- tor, Classical Quantum Gravity 33, 035010 (2016)

work page 2016
[10]

Cyranoski, Chinese gravitational-wave hunt hits crunch time, Nature 531, 150 (2016)

D. Cyranoski, Chinese gravitational-wave hunt hits crunch time, Nature 531, 150 (2016)

work page 2016
[11]

Z. Luo, Y. Wang, Y. Wu, W. Hu, and G. Jin, The Taiji program: A concise overview, Prog. Theor. Exp. Phys. 2021, 05A108 (2020)

work page 2021
[12]

Ruan, Z.-K

W.-H. Ruan, Z.-K. Guo, R.-G. Cai, and Y.-Z. Zhang, Taiji program: Gravitational-wave sources, Int. J. Mod. Phys. A 35, 2050075 (2020)

work page 2020
[13]

W.-H. Ruan, C. Liu, Z.-K. Guo, Y.-L. Wu, and R.-G. Cai, The lisa–taiji network, Nat. Astron. 4, 108 (2020)

work page 2020
[14]

Barausse, J

E. Barausse, J. Bellovary, E. Berti, K. Holley- Bockelmann, B. Farris, B. Sathyaprakash, and A. Sesana, Massive black hole science with elisa, J. Phys.: Conf. Ser. 610, 012001 (2015)

work page 2015
[15]

Cao, H.-J

Z. Cao, H.-J. Yo, and J.-P. Yu, Reinvestigation of moving punctured black holes with a new code, Phys. Rev. D 78, 124011 (2008)

work page 2008
[16]

K. A. Postnov and L. R. Yungelson, The evolution of compact binary star systems, Living Rev. Relativ. 17, 3 (2014)

work page 2014
[17]

Albrecht and P

A. Albrecht and P. J. Steinhardt, Cosmology for grand uniﬁed theories with radiatively induced symmetry breaking, Phys. Rev. Lett. 48, 1220 (1982)

work page 1982
[18]

Wang, J.-G

Z. Wang, J.-G. Lei, and T. S. Collaboration, Devel- opment and on orbit test of taiji-1 inertial reference, Int. J. Mod. Phys. A 36, 2140008 (2021)

work page 2021
[19]

Zhang, P

H. Zhang, P. Xu, Z. Ye, D. Ye, L.-E. Qiang, Z. Luo, K. Qi, S. Wang, Z. Cai, Z. Wang, J. Lei, and Y. Wu, A systematic approach for inertial sensor calibration of gravity recovery satellites and its application to taiji-1 mission, Remote Sens. 15, 3817 (2023)

work page 2023
[20]

X. Wei, L. Huang, T. Shen, Z. Cai, and J. He, Calibration of the in-orbit center-of-mass of taiji-1, 10 Phys. Rev. D 108, 082001 (2023)

work page 2023
[21]

Z. Hu, P. Wang, J. Deng, Z. Cai, Z. Wang, Z. Wang, J. Yu, Y. Wu, Q. Kang, H. Li, and Y. Zhang, The drag- free control design and in-orbit experimental results of “taiji-1”, Int. J. Mod. Phys. A 36, 2140019 (2021)

work page 2021
[22]

Montenbruck and E

O. Montenbruck and E. Gill, Satellite Orbits: Models, Methods and Applications (Springer, Berlin, 2000)

work page 2000
[23]

Bortoluzzi, M

D. Bortoluzzi, M. Da Lio, R. Dolesi, W. Weber, and S. Vi- tale, The lisa technology package dynamics and control, Classical Quantum Gravity 20, S227 (2003)

work page 2003
[24]

Joﬀre, D

E. Joﬀre, D. Wealthy, I. Fernandez, C. Trenkel, P. Voigt, T. Ziegler, and W. Martens, Lisa: Heliocentric formation design for the laser interferometer space antenna mission, Adv. Space Res. 67, 3868 (2021)

work page 2021
[25]

R. E. Kalman, A New Approach to Linear Filtering and Prediction Problems, J. Basic Eng. 82, 35 (1960)

work page 1960
[26]

Sunahara, An Approximate Method of State Estimation for Nonlinear Dynamical Systems, J

Y. Sunahara, An Approximate Method of State Estimation for Nonlinear Dynamical Systems, J. Basic Eng. 92, 385 (1970)

work page 1970
[27]

R. S. Bucy and K. D. Senne, Digital synthesis of non- linear ﬁlters, Automatica 7, 287 (1971)

work page 1971
[28]

F. Dong, H. Liao, C. Jia, and X. Xia, Cal- ibration and adjustment of center of mass (com) based on ekf during in-ﬂight phase, Sci. China, Ser. E: Technol. Sci. 52, 1446 (2009)

work page 2009
[29]

F. Wang, S. Bettadpur, H. Save, and G. Kruizinga, Determination of center-of-mass of gravity recovery and climate experiment satellites, J. Spacecr. Rockets 47, 371 (2010)

work page 2010
[30]

Huang, S

Z. Huang, S. Li, L. Cai, D. Fan, and L. Huang, Estima- tion of the center of mass of grace-type gravity satellites, Remote Sens. 14 (2022)

work page 2022