Inverse Covariance Intersection Fusion CKF algorithm based on Joint Optimization

For nonlinear multi-sensor systems with unknown cross-covariances, an inverse covariance intersection fusion Cubature Kalman filtering algorithm based on joint optimization (genetic algorithm and BP network) is proposed. Firstly, each local estimate is obtained through the Cubature Kalman filter (CKF). And then combined with the batch inverse covariance intersection (BICI) fusion algorithm, the BICI-CKF algorithm is proposed, which can effectively solve the fusion estimation problem of nonlinear system with unknown cross-covariances. Finally, in order to obtain the optimal weighting coefficients of the BICI-CKF algorithm, a joint optimization method based on genetic algorithm and BP network is proposed. This method can not only improve the fusion accuracy, but also significantly increase the fusion speed. The simulation example verifies the effectiveness of the algorithm.