Abstract: With the breakthrough progress of the satellite gravity measurement technology, the demand for the spacecraft's experimental environment increases, and the residual disturbance of the spacecraft must be reduced as far as possible.As the cusped field thruster of China's future alternative gravity thrusters for measuring satellites, the control accuracy of thrusters directly determines the measurement accuracy. A dragfree controller with state mode was designed by the PID method. The controller performs well in predicting the performance of the drag coefficient, the reference mass, the displacement difference and the speed difference of the satellite body. When dealing with satellite operation sudden situation it showed strong stability. However, the PID parameters didn′t reach the optimal solution. Based on this, the control accuracy of the model was optimized. The genetic algorithm was used to filter the PID control parameters. The results show that the control accuracy of the cutoff thruster has improved, the speed and displacement errors in the NTW direction are reduced, the thrust resistance is significantly reduced, the control accuracy is improved, and the requirement is better satisfied.

Key words: cusped field thruster, PID control, genetic algorithm, control accuracy