Research On The Autonomous Navigation Of The Earth Synchronous Orbital Satellite

In this paper, the relevant issues about the absolutely autonomous navigation technologies of earth synchronous orbit satellites are solved. This technology, which has important strategic significance, makes satellites exist independently and not rely on ground equipment. The navigation algorithms are designed and are mainly relied on the information of sun, moon and earth. The navigation algorithms’ feasibility is validated in DSP.First of all, the developments of satellite autonomous navigation technologies and DSP technologies are summarized. A precise orbital dynamic model is established, considering various perturbations.In GTO, when to transfer and how the sensors installed are designed in order to ensure that the sun and the moon are visible in the whole period of GTO. By the output of the sensor, and incorporating the waning and flattening of the earth, the direction vectors from the satellite to sun or moon or earth are determined. The orbital dynamic model is regarded as state equations. The Angle cosine between the direction vector from satellite to earth and the direction vector from satellite to the sun and the Angle cosine between the direction vector from satellite to earth and the direction vector from satellite to the moon are regarded as observed quantities to establish measurement equations. Thus the kalman filter of sun-earth-moon navigation is established. The initial parameters of autonomous navigation should be determined independently, so a method about how to determine initial orbital parameters is put forward.Inertial navigation is used when the satellite enters GEO, so the INS should be corrected online before entering GEO in order to have a relatively accurate inertial navigation. Calculating models of strap down inertial navigation system and the error propagation equations are deduced in the J2000 coordinate system. The state equations of online correction’s kalman filter are established based on the error propagation equations. The deviation values between the velocity and the attitude from inertial navigation and the velocity and the attitude from the sun-earth-moon navigation of GTO are regarded as observed quantities to establish measurement equations of online correction’s kalman filter. When the satellite maneuvers, the initial position and velocity are determined by the navigation of GTO and the initial attitudes are determined by online correction, at the same time, the inertial device’s measuring data should be modified. At GEO, sun-earthmoon navigation is also used. But the satellite will be under four different conditions which are the sun and moon both visible, the sun and moon invisible, only the sun invisible and only the moon invisible. In order to ensure the navigation precision, a star sensor is added. Aiming at different condition, different measurement equation of GEO kalman filter is established.Finally, all the navigation algorithms are analyzed in the numerical simulation and results show that the navigation algorithms can meet the accuracy requirements. In order to ensure navigation algorithm has feasibility in the engineering, they are realized in DSP. Simulator is SEED- XDS510 PLUS and chip is SEED- DEC138. The simulation results in DSP show that the navigation algorithms have enough accuracy and real-time performance.