This paper considers the problem of dynamic modeling and identification of robot manipulators with respect to their elasticities. The so-called flexible joint model, modeling only the torsional gearbox elasticity, is shown to be insufficient for modeling a modern industrial manipulator accurately. Another lumped parameter model, called the extended flexible joint model, is therefore used to improve the model accuracy. In this model, nonactuated joints are added to model the elasticity of the links and bearings. The unknown elasticity parameters are estimated using a frequency-domain gray-box identification method. The conclusion is that the obtained model describes the movements of the motors and the tool mounted on the robot with significantly higher accuracy than the flexible joint model. Similar elasticity model parameters are obtained when using two different output variables for the identification, the motor position and the tool acceleration.
Parameter Estimation of Joint Models Using Global Optimization
