Model Transformation for Enhanced Parameter Identification of Linear Dynamic Systems

System dynamics identification has an important role in engineering, such as for modeling, simulation of dynamic mechanisms and controller design. The accuracy of estimation certainly depends on how the input variables used for estimation are obtained. Often higher order derivatives of the measured independent variables suffer from noise and quantization error compromising the estimation accuracy and conversion. In this paper, we propose a method to avoid successive numerical differentiation for enhanced identification. The proposed method is evaluated using the Least Squares identification method through simulations of twenty sets of dynamic parameters and experiments on a single degree-of-freedom platform. The performance is evaluated in terms of parameter convergence and output prediction.