Predicting a Hybrid System ∗

Predicting the future state of a hybrid system is made exceptionally difficult by uncertainty in the current operating mode of the plant and the possibility of a plant modal transition during the interval over which a prediction is to be made. Unjustified approximations and errors in the system model typically create large errors in the prediction and, for some applications even worse, yield an unrealistic estimate of prediction error covariance. The effect can be an unreasonably optimistic appraisal of the accuracy of a prediction. In this paper, we document the progress to date of an investigation into hybrid prediction using the recently developed Gaussian Wavelet Estimator. We show that the path-length 2 version of this sophisticated estimator can provide consistent predictions and error covariance estimates when applied to an example ship defense problem.