Kalman Filtering in

Reproducing Kernel Hilbert Space

Project Description:

This project studied the feasibility of a (nonlinear) mapping to a linear functional space (a Reproducing Kernel Hilbert Space) to implement the Kalman filter equations and achieve performance commensurate with nonlinear models. One of the difficulties of this approach investigated is the growing memory requirements that will be dealt with novel sparsification criteria and algorithms based on information theory. The developed approaches are applied in automotive engine control, and state estimation in non-linear and non-Gaussian systems.

Peer-Reviewed Publications:

  • Pingping Zhu, Badong Chen, Jose C. Principe, "Learning Nonlinear Generative Models of Time-Series with a Kalman Filter in RKHS", IEEE Trans. Signal Processing, Vol. 62 (1): 141-155, 2014. [Link]
  • Pingping Zhu, Jose C. Principe, "Kernel Recurrent System Trained by Real-Time Recurrent Learning Algorithm", Acoustics, Speech, and Signal Processing (ICASSP), International Conference on, 2013. [Link]
  • Pingping Zhu, Jose C. Principe, "Analysis on Extended Kernel Recursive Least Squares Algorithm", International Joint Conference on Neural Networks (IJCNN), 2013. [Link]
  • Pingping Zhu, Badong Chen, Jose C. Principe, "A Novel Extended Kernel Recursive Least Squares Algorithm", Neural Networks, Vol.32,349-357, 2012. [Link][PDF]
  • Pingping Zhu, Badong Chen, Jose C. Principe, "Extended Kalman Filter Using a Kernel Recursive Least Squares Observer", International Joint Conference on Neural Networks (IJCNN), pp. 1402-1408, Jul. 2011. [Link]