Stellar sensor based nonlinear model error filter for gyroscope drift extraction

The conventional gyroscope based spacecraft attitude determination approaches are afflicted by the integration drift of gyroscope. To ameliorate the performance of gyroscope, many compensation algorithms have been developed. A novel spacecraft attitude determination algorithm, which is based on the stellar sensor and nonlinear model error filter algorithm, is discussed in this paper. This algorithm uses the attitude quaternion obtained from stellar sensor to generate the attitude information of spacecraft, which could compensate the drift error of the gyroscope unit. This approach could be utilized both as the gyroscope error compensation in high precision integrate navigation system and as an attitude determination unit individually in low precision mini-satellite task. Two verifications of different satellite orbits have been launched and the performance of our approach was proved.