Cartesian space neural network control of robot manipulators

• Autorzy: Moise A. , Boutalis Y.S. , Mertzios B.G.
• Języki publikacji: EN

Abstrakty

EN

It is very common that manipulators are subject to structured and/or unstructured uncertainties. Structural uncertainty is characterised by having a correct dynamical model but with parameter uncertainty due to imprecision of the manipulator link properties, unknown loads, inaccuracies in the torque constants of the actuators, and so on. Unstructured uncertainty is characterised by unmodelled dynamics. Neural Network (NN) controllers are usually introduced to generate additional inputs to compensate for disturbances due to model uncertainties. It is clear that the higher the degree of nonlinearity exists in the uncertainties, the greater benefits neural networks (NNs) can contribute. In Cartesian space control, more robust control is required due to the following problems: (i) the inverse dynamic in Cartesian space is more complicated than that in joint space; (ii) the singularity of Jacobian (J) becomes problem at hand when positions in Cartesian space are calculated from joint measurements; - more uncertainties are present. In this paper, we examine some of these issues by studying the Cartesian space robot manipulator control problem. Although NN compensation for model uncertainties has been traditionally carried out by modifying the joint torque/force of the robot, it is also possible to achieve the same objective by using the NN to modify the reference Cartesian trajectory. We present four different NN controller designs to achieve disturbance rejection for an uncertain system.

Słowa kluczowe

EN

robustness control systems; two term control systems; uncertain systems

• Wydawca: Oficyna Wydawnicza Politechniki Wrocławskiej
• Czasopismo: Systems Science
• Rocznik: 2001
• Tom: Vol. 27, nr 4
• Strony: 35--46
• Opis fizyczny: Bibliogr. 10 poz.,

Twórcy

autor

Moise A.

autor

Boutalis Y.S.

autor

Mertzios B.G.

• Petroleum-Gas University of Ploiesti, Dept. of Automatic Control and Comp., Bd. Bucuresti, Nr. 39, 2000 Ploiesti, Prahova, Romania