Satellite high-precision combined attitude determination method based on extreme learning machine network compensation

The invention belongs to the technical field of remote sensing mapping, and discloses a satellite high-precision combined attitude determination method based on extreme learning machine network compensation. The method comprises the following steps: constructing a system state equation according to a satellite attitude kinematics model and a gyro error model; constructing a system measurement equation according to the star sensor error model; estimating attitude parameters by adopting an AKF model; constructing a data set according to the filtering process quantity and the filtering result, and randomly selecting a part of the data set as a training set; building an extreme learning machine network model, and performing off-line training on the extreme learning machine network by using thetraining set to obtain network parameters; setting different star sensor measurement error parameters, inputting AKF filtering process quantity into a trained extreme learning machine network to obtain attitude parameter com