A new integrated algorithm for spacecraft attitude determination

Global Positioning System (GPS) /Inertial Navigation System (INS) integrated system is continuously gaining research interest in many positioning and navigation fields for space application. Kalman filtering is one of the best methods for multi-sensor integration, but, it has some drawbacks on stability, computation load, robustness, and system observability performances. Meanwhile, Kalman filters perform adequately only under some certain dynamic models. In this paper, a new GPS/INS integration filtering algorithm based on neural network is studied. One of neural network-based adaptive Kalman filter is used for an integration scheme for the Attitude Determination GPS/INS integrated navigation system. To assess the performance of the proposed integrated model, some simulation is done to compare against the traditional Kalman filter model. The results indicate that the proposed model provides a significant improvement in some performance, such as accuracy, stability, robustness, and so on.