Experimental Robot Inverse Dynamics Identification Using Classical and Machine Learning Techniques

Vinzenz Bargsten, José de Gea Fernández, Yohannes Kassahun

In: International Symposium on Robotics. International Symposium on Robotics (ISR) June 21-22 München Germany 2016.


This paper shows the experimental identification of the inverse dynamics model of a KUKA iiwa lightweight robot. We use experimental data from optimal identification experiments to evaluate and compare two different identification approaches: a classical method using a parametrized robot dynamical model and a machine learning method. Both methods accurately estimate the dynamics model and this paper will discuss the pros and cons of each method.


ISR16_Dynamics_Identification.pdf (pdf, 932 KB )

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz