Non-gravitational force measurement and correction by a precision inertial sensor of TianQin-1 satellite

Non-gravitational force models are critical not only for the applications of satellite orbit determination and prediction, but also for the studies of gravitational reference sensors in space-based gravitational wave detection missions and accelerometers in gravity satellite missions. In this paper, based on the inertial sensor data from the TianQin-1 (TQ-1) mission, a correction has been made in the non-gravitational force models by applying additional terms related to the orbital periods. After taking into account this correction, about 37 hours of TQ-1 inertial sensor data is calibrated in the sensitive axes, i.e. y- and z-axes, by comparing with the simulated non-gravitational accelerations. It is indicated that the peak-to-peak value of the non-gravitational acceleration correction terms are about 2% and 13% of the measured accelerations in the y- and z-axes, respectively. Within the frequency band below 0.01 Hz, the root mean square of calibration residual errors in y- and z-axes are suppressed from 1.03 x 10(-9) and 3.872 x 10(-9) m s(-2) to 8.14 x 10(-10) and 1.343 x 10(-9) m s(-2), respectively. The bias and scale factor of the inertial sensor are also obtained from the calibration by the method of least-squares fit. Meanwhile, the inertial sensor measurements are validated and their signal compositions are analyzed.

Automated summary

Non-gravitational force models play a critical role in satellite orbit determination and prediction, as well as in gravitational reference sensors and accelerometers in space missions. In this study, a correction was made in the non-gravitational force models based on data from the TQ-1 mission, resulting in a significant improvement in the calibration of inertial sensor data. The peak-to-peak value of the non-gravitational acceleration correction terms and the root mean square of calibration residual errors were both reduced, while the bias and scale factor of the inertial sensor were obtained through least-squares fit calibration. Additionally, the validation and analysis of signal compositions of the inertial sensor measurements were conducted.