Model Structure Selection for Manipulator Trajectory Tracking

This paper outlines a method of choosing a model structure which can describe the dynamics of a manipulator. The model structure is based on a "Black Box" approach, therefore the model does not attempt to determine physical parameters such as moments of inertia or link masses. The model structure is used to estimate parameters which best fit the selected model structure to the observed input/output data sequence. The manipulator dynamics are written as a fourier series expansion. The fourier coefficients are identified by a least squares algorithm. This paper will present a parameter selection criteria which will determine the model structure one should choose. The tracking accuracy of controllers based on different model structures will be compared. The model structures are chosen based on a parameter selection criteria. It is demonstrated that simple model structures can be used to model the nonlinear dynamics of a robotic manipulator.