Dual-Band Infrared and Geomagnetic Fusion Attitude Estimation Algorithm Based on IMMEKF

In order to further improve the performance of dual-band infrared and magnetic fusion attitude measurement method, an interactive multimodel extended Kalman filter (IMMEKF) attitude fusion algorithm is proposed to calculate the rotating missile's full attitude angle by using the data collected from dual-band two-axis infrared sensors and three-axis geomagnetic sensor. Four filtering models are used to solve the problem of inaccuracy in the single-model filtering process. The interactive multimodel and extended Kalman filter attitude estimation algorithms are combined by the probability density function and are used to update the model probability in real time. Each model's probability weights are modified to achieve the optimal attitude angle estimation according to the environment. The semiphysical experimental results show that the IMMEKF algorithm has higher accuracy and reliability compared with other latest state-of-the-art fusion estimation algorithms under normal infrared and geomagnetic conditions. In the case of infrared or geomagnetic anomalies, only the IMMEKF algorithm can avoid using the affected sensor data and calculate an accurate value of attitude angle. The reliability of the dual-band infrared and geomagnetic combined attitude estimation method is improved, which provides an effective technical implementation for the rotating missile attitude measurement.