Observability of measurement bias for nonlinear systems arising from satellite autonomous navigation and attitude determination

Paper number

IAC-05-C1.P.25

Author

Dr. Yong Li, China Academy of Space Technology (CAST), China

Coauthor

Mr. Liang Dong LIU, China Academy of Space Technology (CAST), China

Coauthor

Mrs. Chun-qing Zhang, China Academy of Space Technology (CAST), China

Year

2005

Abstract

Future missions of aerospace including satellite precise target pointing, satellite constellations and formation flying will require accurate and up-to-date knowledge of satellite orbit and attitude parameters. Determination of the satellite position, velocity and attitude are done in a number of ways using a variety of measurements. Autonomous onboard navigation and attitude determination can be preferred in order to minimize the costly ground segment and enhance satellite independence with respect to ground station unavailability. Kalman filter is the most widely used filter for satellite autonomous navigation and attitude determination. Lots of analysis and simulation demonstrated that the effect in the accuracy of navigation and attitude determination due to the measurement noise is effectively diminished, while the systematic errors may lead to notable errors between state estimations and true states. In practice, the measurements of sensors are corrupted by errors from various sources which may be systematic or stochastic. Therefore, the key problem for improving the performance of autonomous navigation and attitude determination is how to precisely estimate and calibrate systematic errors in the measurements, while the precondition for precise estimation of the systematic errors is that these errors are all observable. In this paper, nonlinear systems with measurement constant bias arising from satellite autonomous navigation and attitude determination are concerned. Sufficient conditions and necessary conditions for the observability of system states, as well as all measurement constant bias are derived. A satellite autonomous navigation system using Earth, Sun and Moon sensors, and a satellite attitude determination system using gyroscope, Earth and Sun sensors are investigated respectively. It is shown that under certain assumptions the system states, as well as measurement constant bias are all observable even in the case that all measurement are corrupted by constant bias errors. Results presented in this paper provide the theoretical foundation for design of self-calibration filter in satellite autonomous navigation and attitude determination. We give our research results on design of self-calibration filter and its application in a satellite autonomous navigation system in the paper “Application of Self-Calibration Filter for Satellites Autonomous Orbit Determination”, which also submitted to IAC 2005.
We would like to indicate that the material in this paper is new and original, and the paper was not presented at a previous meeting. The authors were supported by the National Natural Science Foundation of China, therefore the financing and attendance of an author at the respective IAC at Fukuoka to present the paper is assured.

Abstract document

IAC-05-C1.P.25.pdf

Manuscript document

IAC-05-C1.P.25.pdf (🔒 authorized access only).

To get the manuscript, please contact IAF Secretariat.