Parameter and Delay Estimation of Continuous-time Models from Irregularly Sampled Output

Linear filter approach might be the most commonly used method for continuous-time identification. Recently we have proposed a new linear filter method for simultaneous estimation of time delay and other parameters of continuous-time models in (Ahmed et al., 2006). The proposed method involves choice of filter parameters and the filter structure is restricted to all real pole form. In this paper, the linear filter method is presented for a general structure of the filter. Also the filter parameters are updated iteratively. Next, an algorithm is prescribed to make this method applicable when the output is sampled irregularly. To demonstrate the performance of the proposed algorithm, results from a simulation example as well as an experimental example are presented. Copyright c ©2006 IFAC.