Multi-source information fusion for space gravitational waves detection thrust estimation on orbit

Space Gravitational Wave (GW) detection task puts forward super-low thrust noise for micro-newton propulsion system. Study shows thrust noise can be reduced by thrust closed loop feedback control structure. The effect is determined by thrust estimation accuracy. Traditionally, thrust estimation is realized by electric parameters and thrust estimation equation. However, this way is faced with high thrust estimation error in the low frequency band. Besides, the displacement sensor information of the drag-free control system can also be used for thrust estimation via disturbance observer. However, this way is only effective for low-frequency thrust due to the measurement noise of displacement sensor. A data fusion method is proposed in this paper to enhance thrust estimation accuracy. In this method, the high frequency thrust noise is estimated by electric parameters and thrust estimation equation. Meantime, the low frequency thrust noise is estimated by displacement sensor information and disturbance observer. In this paper, the cusped field thruster is used as the actuator of drag-free control system to simulate. Simulation results show thrust estimation by data fusion can effectively reduce thrust noise in the middle frequency band and enhance stability of satellites compared with that by electric parameters.