In order to overcome the shortcomings that traditional least square algorithm tends to become divergence with the increase of matching points INS/SMNS (inertial navigation system/scene matching navigation system) integrated navigation system, an improved LAF(Lyapunov adaptive filtering), based on MIMO (multi-input-multi-output) system, is presented. On the basis of the information of observation formula in the algorithm, the recursive formula of LAF is obtained by the definition of filtering error and priori estimation error of LAF. The stability and convergence of LAF are proved by Lyapunov stability theory. LAF is applied to INS/SMNS integrated navigation system, and the simulation results demonstrate that compared with RLS( recursive least square) algorithm, LAF is simpler and more reliable, and its filtering performance is better.
Jiang Chunhong, Chen Zhe. Multisensor fusion using Hopfield neural network in INS/SMGS integrated system . Proceedings of IEEE 6th International Conference on Signal Processing . Beijing:IEEE, 2002. 1199~1202
�� ��. ����ϵ���ϵͳ�㷨�о���������� . ����:�������պ����ѧ,�Զ�����ѧ���������ѧԺѧλ����,2000. 84~99 Min Hu. Research of the algorithms and development of the software in INS/GPS/TAN/SMN integrated navigation system . Beijing:School of Automation Science and Electrical Engineering, Beijing University of Aeronautics and Astronautics,2000. 84~89(in Chinese)
Spripada N R, Fisher D G. Improved least square identification [J].International Journal of Control.1987,46(6):1889-1913
Tzuu-Hseng S Li, Chyi-Cherng Lai. Lyapunov function based fuzzy state estimator . IEEE International Symposium on Circuits and System . Seattle:IEEE, 1995. 497~500
Sneg Kah Pjooi, Man Zhihong, Wu H R. Nonlinear active noise control using Lyapunov theory and RBF network . Proceeding of the IEEE Signal Processing Society Workshop Neural Networks for Signal Processing X . Piscataway:Institute of Electrical and Electronics Engineers Inc, 2000. 916~925
Sneg Kah Pjooi, Man Zhihong, Wu H R.null Lyapunov-theory-based radial basis function networks for adaptive filtering[J]. IEEE Transaction on Circuits and System-Ι:Fundamental Theory and Applications.2002,49(8):1215-