A parameter estimation approach to state observation of nonlinear systems

A novel approach to the partial state estimation problem is proposed. Instrumental to this approach is the reformulation of the state estimation problem as a problem of estimation of a constant parameter related to the systems initial conditions. The admissible class of the systems is defined via two assumptions related to solvability of a partial differential equation and our ability to estimate the unknown parameters. The proposed approach is shown to be applicable to position estimation of a class of electromechanical systems.