State and parameter estimation via discrete-time exponential forgetting factor observer
Résumé
The paper presents the discrete-time counterpart of the continuous-time exponential forgetting factor observer. We obtain a Kalman-like observer to the extent that it relies on the uniform complete observability property specific to LTV systems and can be used to handle the estimation of state-afine systems. However, it does not require the complete uniform controllability property like the Kalman observer does. Like any other state observer it can be used for parameter estimation if the eventual unknown parameters are turned into state variables. This use of the observer is illustrated through an example of state and parameter estimation in the case of an AC machine.