Nonlinear parameter estimation using polynomials and resultants - Application to electrical drives

This paper proposes the use of algebraic geometry tools, i.e. resultants, for the estimation of uncertainties in a nonlinear system. With the focus on the case-study of electrical drives, an algorithm based on the solution of polynomial equalities obtained by processing the system outputs is shown to be capable to accurately reconstruct the motor resistance and initial angular position in finite time using only the measurement of the currents. Moreover, the procedure to extend the method for handling more challenging scenarios or dealing with other types of systems is provided.