System identification using the neural-extended Kalman filter for state-estimation and controller modification

• Autorzy: Stubberud S. C. , Kramer K. A.
• Języki publikacji: EN

Abstrakty

EN

The neural extended Kalman filter (NEKF) is an adaptive state estimation technique that can be used in target tracking and directly in a feedback loop. It improves state estimates by learning the difference between the a priori model and the actual system dynamics. The neural network training occurs while the system is in operation. Often, however, due to stability concerns, such an adaptive component in the feedback loop is not considered desirable by the designer of a control system. Instead, the tuning of parameters is considered to be more acceptable. The ability of the NEKF to learn dynamics in an open-loop implementation, such as with target tracking and intercept prediction, can be used to identify mismodeled dynamics external to the closed-loop system. The improved system model can then be used at given intervals to adapt the state estimator and the state feedback gains in the control law, providing better performance based on the actual system dynamics. This new approach to neural extended Kalman filter control operations is introduced in this paper using applications to the nonlinear version of the standard cart-pendulum system.

Słowa kluczowe

EN

system identification; adaptive control; neural networks; Kalman filters

• Wydawca: Oficyna Wydawnicza Politechniki Wrocławskiej
• Czasopismo: Systems Science
• Rocznik: 2008
• Tom: Vol. 34, no 2
• Strony: 17--25
• Opis fizyczny: Bibliogr. 16 poz., wykr.

Twórcy

autor

Stubberud S. C.

autor

Kramer K. A.

• Rockwell-Collins, 12365 First American Way, Poway, CA, USA 92064,

Bibliografia

• [1] Santina M.S., Stubberud A.R., Hosteller G.H., Digital control system design, 2nd ed., Fort Worth, Texas, Saunders College Publishing, 1994.
• [2] Rhodes I.E., A tutorial introduction to estimation and filtering, IEEE Trans. Automatic Control, Vol. AC-16, No. 6, December 1971, pp. 688-706.
• [3] Ljung L, System identification: theory for the user, Englewood Cliffs, NJ, Prentice-Hall, 1987.
• [4] Narendra K.S., Annaswamy A.M., Stable adaptive systems, Prentice Hall, Englewood Cliffs, New Jersey, 1989.
• [5] Stubberud S.C., Lobbia R.N., Owen M., An adaptive extended Kalman filter using artificial neural networks, International Journal on Smart System Design, Vol. 1, 1998, pp. 207-221.
• [6] Stubberud S.C., Lobbia R.N., Owen M., An adaptive extended Kalman filter using artificial neural networks, Proc. 34th IEEE Conf. Decision and Control, New Orleans, Louisiana, 1995, pp. 1852-1856.
• [7] Stubberud A.R. A validation of the neural extended Kalman filter, Proc. 2006 Int. Conf. Systems Engineering, Coventry, UK, September 2006, pp. 3-8.
• [8] Stubberud S.C., Owen M., Controller modification using improved models of the neural extended Kalman filter, Proc. 14th Int. Conf. Systems Engineering, Coventry, UK 2000, pp. 519-523.
• [9] Owen M.W., Stubberud A.R., A neural extended Kalman filter multiple model tracker, Proc. OCEANS 2003, San Diego, California, September 2003, pp. 2111-2119,
• [10] Stubberud S.C., Kramer K.A., Predictive guidance intercept using the neural extended Kalman filter tracker. Control and Intelligent Systems, Vol. 35, No. 3, 2007, pp. 228-235.
• [11] Haykin S., Kalman Filtering and Neural Networks, Wiley Interscience, New York 2001
• [12] Iglehart S., Leondes C., Estimation of a dispersion parameter in discrete Kalman filtering, IEEE Trans. Automatic Control, Vol. 19 (3), 1974, pp. 262-263.
• [13] White H., Stinchcombe M., Universal approximation using feedforward networks with non-sigmoid hidden layer activation functions, Proc. IEEE Int. Joint Conference on Neural Networks, Washington, D.C., 1989, pp. 613-617.
• [14] Singhal S., Wu L., Training multilayer perceptrons with the extended Kalman algorithm. Advances in Neural Processing Systems I, D.S. Touretsky, ed., Morgan Kaufmann, 1989, pp. 133-140.
• [15] Goodwin G.C., Graebe S.F., Salado M.E., Control system design. Prentice Hall, Upper Saddle River, NJ, 2001.
• [16] Stubberud S.C., Kramer K.A., State estimation and controller modification using an improved model generated by a neural extended Kalman filter, Proc. 16th Int. Conf. System Science, Vol. 1, Wrocław, Poland, September 2007, pp. 470-480.