Physical parameter estimation of the nonlinear dynamics of a single link robotic manipulator with flexible joint using the HMF method

Abstract

The application of the Hartley modulating functions (HMF) method is investigated to estimate the physical parameters of a single link robotic manipulator with flexible joint. The approach uses a weighted least-squares algorithm in the frequency domain. Knowing the structure of a continuous-time system, the identification method will only focus on the estimation of the physically-based system parameters using input and noise-corrupted output signal records. The methodology facilitates the conversion of a system differential equation into an algebraic equation in the parameters. Numerical simulations for a single link robotic manipulator with flexible joint are reported, which illustrate the application and performance of the methodology. The HMF method shows promising results for the identification of physically-based continuous-time nonlinear systems in the presence of noticeable measurement noises.