论文信息 - On satisfying strict positive real condition for convergence by over-parameterization

On satisfying strict positive real condition for convergence by over-parameterization

Strict Positive Real (SPR) conditions for global convergence of stochastic adaptive control and recursive identification schemes have appeared in a number of analyses using different approaches. It is shown that this Strict Positive Real condition on the unknown operator associated with the plant can be removed in every case by a sufficient over-parameterization of the parameter vector.

G. Franklin | S. Shah

[1] H. Kushner. Convergence of recursive adaptive and identification procedures via weak convergence theory , 1977 .

[2] G. Goodwin,et al. An adaptive d-step ahead predictor based on least squares , 1980 .

[3] Lennart Ljung,et al. On positive real transfer functions and the convergence of some recursive schemes , 1977 .

[4] P. Ramadge,et al. Discrete Time Stochastic Adaptive Control , 1981 .

[5] A. Benveniste,et al. Analysis of stochastic approximation schemes with discontinuous and dependent forcing terms with applications to data communication algorithms , 1980 .