Globally optimal bounding ellipsoid algorithm for parameter estimation using artificial neural networks

This paper develops a real-time implementation of a globally optimal boundi ng ellipsoid (GOBE) algorithm for parameter estimation of linear-in-paramet er models with unknown but bounded (UBB) errors. A recently proposed recurs ively optimal bounding ellipsoid (ROBE) algorithm is introduced and a GOBE algorithm is derived through repeating this ROBE algorithm. An analogue art ificial neural network (ANN) is provided to implement the GOBE algorithm in real time. Convergence analyses on the ROBE, the GOBE algorithms, and the analogue ANN implementation of the GOBE algorithm are presented. No persist ent excitation condition is required to ensure the convergence. Simulation results show the good performances of these algorithms and the ANN implemen tation.