Eric Wan, Rudolph van der Merwe, Alex Nelson
Dual estimation refers to the problem of simultaneously estimating the state of a dynamic system and the model which gives rise to the dynam(cid:173) ics. Algorithms include expectation-maximization (EM), dual Kalman filtering, and joint Kalman methods. These methods have recently been explored in the context of nonlinear modeling, where a neural network is used as the functional form of the unknown model. Typically, an ex(cid:173) tended Kalman filter (EKF) or smoother is used for the part of the al(cid:173) gorithm that estimates the clean state given the current estimated model. An EKF may also be used to estimate the weights of the network. This paper points out the flaws in using the EKF, and proposes an improve(cid:173) ment based on a new approach called the unscented transformation (UT) . A substantial performance gain is achieved with the same order of computational complexity as that of the standard EKF. The approach is illustrated on several dual estimation methods.