Identification of a robotic arm using optimization methods for model estimation

ABSTRACT I I The system identification procedure is a powerful andflexible tool for the modeling of dynamic systems. Thispaper implements the theory of parametric identlfidatlon in order to estimate a valid model of a flexible roboticarm. For this purpose experimental data Is used for the,stimation of ARMAX SISO mocjels. A two-stagesIdentification procedure (non-parametric & parametric)provides an Insight about the system underidentification. In the first stage, known signal analysismethods are applied (correlation-spectral analysis) forthe estimation of frequencies and frequency response.and in the second stage, the estimation of ARMAXmodels is performed in order to fit a transfer functionmodel to collected input-output data set. For theestimation of model's coefficients, use of EvolutionaryAlgorithms is implemented. l INTRODUCTIONThe essential concept behind System Identification relies on the simple principle that itis impossible to sit ona desk and figure out how the world works: italso has tobe studied. In an engineering point of view this meansthat. during a modeling process, the use of experimental