An indirect method for transfer function estimation from closed loop data

The CONTSID toolbox for Matlab: extensions and latest developments

“…Therefore, rather simpler but sub-optimal two-stage SRIVC-based approaches have been very recently developed Young et al [2009]. These are in the same spirit as the two-stage algorithm suggested by Van den Hof and Schrama [1993] for discrete-time systems and they have the same advantage of not requiring prior knowledge of the control system. However, the new twostage algorithms are more sophisticated than the Van den Hof and Schrama approach because each stage exploits the appropriate open loop SRIVC/RIVC algorithms: namely, the SRIVC algorithm (rather than the FIR model estimation used by Van den Hof) for estimating the control input signal, followed by the estimation of the enclosed, controlled system, based on this estimated control input, using either the SRIVC or the full RIVC algorithm.…”

Identification and parameter-varying decoupling of a 3-DOF platform with manipulator

“…[6], [7]), instrumental variable ( [8], [9]) or direct identification ( [10], [11]). Recently developed methods are for example two-stage [12] or dual-Youla parameterization (e.g. [13]).…”

Closed-loop identification with a quantizer

“…Three branches exist within the joint input-output approach. They are the coprime factor identification scheme, the two-stage method [3] or the two-step method [10] and the projection method [8]. Esmaili et al [11] discussed the asymptotic and finite data behavior of some closed-loop identification methods while Gevers et al [12] derived the asymptotic variance expressions for models that are identified on the basis of closed-loop data, showing that direct and indirect methods lead to the same asymptotic variance.…”

“…The quantizer is used to generate a persistently exciting signal so that the two-stage method by Van den Hof and Schrama [3] can be used directly for identification.…”