Adaptive Filtering based on Recurrent Neural Networks

c Abstract— Kalman filter is an optimal filtering solution in certain cases, however, it is more often than not, regarded as a non-robust filter. The slight mismatch in noise statistics or process model may lead to large performance deterioration and the loss of optimality. This research paper proposes an alternative method for robust adaptive filtering concerning lack of information of noise statistics. The method is based on the application of recurrent neural networks trained by a dynamic identity observer. The method is explained in details and tested in the case analysis of object tracking model. Performance evaluation is made for cases of the standard Kalman filter, a noise-adaptive Kalman filter, the adaptive filter with a recurrent neural network trained by a static identity observer, and the adaptive filter with recurrent neural network trained by a dynamic identity observer. The results for different noise statistics as well as noise statistics mismatches are compared and presented. It is shown that in cases with a lack of knowledge of the noise statistics it is beneficial to use the filtering method proposed in this research work.

[1]  Jong-Koo Park,et al.  The concept and design of dynamic state estimation , 1999, Proceedings of the 1999 American Control Conference (Cat. No. 99CH36251).

[2]  R. E. Kalman,et al.  A New Approach to Linear Filtering and Prediction Problems , 2002 .

[3]  Tony T. Lee.  A direct approach to identify the noise covariances of Kalman filtering , 1980 .

[4]  Dan Simon,et al.  Hybrid Kalman / Minimax Filtering in Phase-Locked Loops , 1996 .

[5]  S. Haykin,et al.  Adaptive Filter Theory , 1986 .

[6]  Keith Glover,et al.  Robust control design using normal-ized coprime factor plant descriptions , 1989 .

[7]  R. Stengel.  Stochastic Optimal Control: Theory and Application , 1986 .

[8]  Guanrong Chen,et al.  Kalman Filtering with Real-time Applications , 1987 .

[9]  H. W. Sorenson,et al.  Kalman filtering : theory and application , 1985 .

[10]  Pramod P. Khargonekar,et al.  A Comparison of Classical Stochastic Estimation and Deterministic Robust Estimation , 1991, 1991 American Control Conference.

[11]  H. Kushner.  Introduction to stochastic control , 1971 .

[12]  T. Soong,et al.  Mathematics of Kalman-Bucy filtering , 1985 .

[13]  Evanghelos Zafiriou,et al.  Robust process control , 1987 .

[14]  Joydeep Ghosh,et al.  Hierarchical adaptive Kalman filtering for interplanetary orbit determination , 1998 .

[15]  M. Kao,et al.  Multiconfiguration Kalman filter design for high-performance GPS navigation , 1983 .

[16]  T. Başar,et al.  A New Approach to Linear Filtering and Prediction Problems , 2001 .

[17]  C. S. George Lee,et al.  Neural fuzzy systems: a neuro-fuzzy synergism to intelligent systems , 1996 .

[18]  D. Sworder,et al.  Introduction to stochastic control , 1972 .

[19]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[20]  B. Anderson,et al.  Optimal Filtering , 1979, IEEE Transactions on Systems, Man, and Cybernetics.

[21]  Zeljko M. Durovic,et al.  Robust estimation with unknown noise statistics , 1999, IEEE Trans. Autom. Control..

[22]  Gary J. Balas,et al.  μ-analysis and synthesis toolbox: for use with Matlab , 1994 .

[23]  Robert R. Tenney,et al.  A tracking filter for maneuvering sources , 1977 .

[24]  Peter S. Maybeck,et al.  Multiple-model adaptive estimation using a residual correlation Kalman filter bank , 2000, IEEE Trans. Aerosp. Electron. Syst..

[25]  Maciejowsk.  Multivariable Feedback Design , 1989 .