New developments in state estimation for nonlinear systems (Q5925916)

The authors consider the nonlinear model with state \(x\) and observation \(y\): \[ x_{k+1}= f(x_k, v_k),\quad y_k= g(x_k, w_k), \] where \(v_k\) and \(w_k\) are assumed i.i.d. and independent of current and past states. They derive state estimators based on interpolation formulas and show that the estimators perform better than the well-known filters (such as the extended Kalman filters) based on Taylor approximations.