Mar. 31, 1999 to Apr. 3, 1999
Markus Krabbes , Otto-von-Guericke-University Magdeburg
Christian Döschner , Otto-von-Guericke-University Magdeburg
A modeling of robot manipulator dynamics by means of a neural architecture is presented. Such model is applicable to generate a decoupling and linearizing feedback in the control system of the robot. In a structured model approach, a RBF-like neural network is used to represent and adapt all model parameters with their dependences on the joint positions.The neural network is hierarchically organized to reach optimal adjustment to the common structural knowledge about the identification problem. A fixed, grid based neuron placement together with application of polynomial basis functions is utilized favorably for a very effective recursive implementation. That way a neural network based online identification of a dynamic model is enabled for a complete industrial 6 joint robot with reasonable effort and good results.
Markus Krabbes, Christian Döschner, "Modeling of Robot Dynamics Based on a Multi-Dimensional RBF-Like Neural Network", ICIIS, 1999, Information, Intelligence, and Systems, International Conference on, Information, Intelligence, and Systems, International Conference on 1999, pp. 180, doi:10.1109/ICIIS.1999.810257