Study and comparison of membership-set outer bounding
暂无分享,去创建一个
When model output-error is unknown but bounded, the set of parameter called the membership-set and consistent with the measurement, the model and the error bounds is a polytope, the complexity of which is increasing with the model order and the measurement number. Such complexity bereaves the algorithm updating the polytope to be used in real time control. To overcome this complexity, the exact polytope is approximated by an outer bounding region with a simpler shape and which approximate it as most as possible. In this paper we present first algorithms which update the exact polytope and its outer bounding approximations such as ellipsoid, parallelotope or orthotope. A comparative study of these algorithms is carried out. The comparison is based on the volume of the final region, the computing and the convergence times, the numerical complexity and the updating rate.
[1] A. Vicino,et al. Sequential approximation of feasible parameter sets for identification with set membership uncertainty , 1996, IEEE Trans. Autom. Control..
[2] Gérard Favier,et al. A Review and a Comparison of Robust Estimation Methods , 1992 .
[3] Gérard Favier,et al. Recursive Determination of Parameter Uncertainty Intervals for Linear Models with Unknown But Bounded Errors , 1994 .
[4] Y. F. Huang,et al. On the value of information in system identification - Bounded noise case , 1982, Autom..