This paper considers the state observation problem for bounded dynamic stochastic systems by using B-spline model to approximate output probability density functions. First, a form of residual signal is considered for the output probability density function using square root B-spline neural network model,and the on-line tuning of the observer gain is obtained using Lyapunov analysis. Then a new logarithm B-spline model is presented and the adaptive observer is designed. Finally, two simulated examples are used to demonstrate the proposed algorithms, and desired results have been obtained.
|Number of pages||8|
|Journal||Journal of the Graduate School of the Chinese Academy of Sciences|
|Publication status||Published - 2005|
- adaptive updating rule
- residual signal
- B-spline model