Adaptive sliding mode controller design for rigid robot manipulators

碩士 === 國立中央大學 === 電機工程研究所 === 83 === In this thesis, we discuss the problem of the path track- ing of rigid robot manipulators.A good path tracking contro- ller can not only depress the influence of the disturbances, which may result from i...

Full description

Bibliographic Details
Main Authors: Chu,Peng-hsien, 朱鵬憲
Other Authors: Prof.Shyu, Kuo-kai
Format: Others
Language:en_US
Published: 1995
Online Access:http://ndltd.ncl.edu.tw/handle/38353658694153596989
Description
Summary:碩士 === 國立中央大學 === 電機工程研究所 === 83 === In this thesis, we discuss the problem of the path track- ing of rigid robot manipulators.A good path tracking contro- ller can not only depress the influence of the disturbances, which may result from internal system uncertainties and ext- ernal disturbace, but also reduce the tracking error. Moreo- ver,the control method must be simple enough to be feasible. It is thus that we,in this thesis, take the advantage of the adaptive control theory to release the assumption that the bounds of the uncertainties and disturbances must be known previously. Furthermore, the controller is combined with the sliding mode control to enhance the robustness of the system to disturbances. At the same time, the adaptive sliding mode controller without serious chattering guarantees that the tracking error convergeto zero, which is different from the result derived by the boundary layer of the conventional va- riable structure control systems. Furthermore,the advantages of the characteristics of the regressor matrix of the robot manipulator dynamic equ ation can be fully ultized in this thesis to avoid the heavy computation load which is needed for the estimation and inversion of the inertia matriix in the inverse dynamics method. Therefore, a new compensator is designed to decrease the influence resulting from the stri- ct assumptions in inverse dynamics method. Accordingly,this thesis is to derive an effective and practical control for robot manipulator. The control effect is demonstrated by si- mulations.