Parameter identification for nonlinear systems

摘要

Preliminary results from an investigation into identification of parameters appearing nonlinearly in system dynamics are presented. After presentation of background material on identifiability of such systems, an identifier based on the extended Luenberger observer is constructed for a class of nonlinear systems. Conditions sufficient for global convergence of its parameter estimates are discussed. An example illustrates the identifier's construction, and computer simulation results depict the parameter estimate's global exponential convergence.