Identification of the dynamic parameters of the planar robot model at 2 degrees of freedom using the least squares method

Abstract

The designation of dynamic parameters of robots is generally based on the use of the inverse model which is linear with respect to the dynamic parameters to be identified. In order to obtain an undetermined system, this model is sampled while the robot is driven with exciting movements. The optimum solution is obtained using the least linear squares. Moreover, the identification is carried out in a closed loop due to the unstable double integrator character of the robots. This paper focuses on this technique and proposes an extension of this method which is applied to a robot with 2 degrees of freedom in the plane.