Robot manipulator control using adaptive-gain sliding modes.

摘要

In this thesis, adaptive sliding mode controls are presented as robust controls for robot manipulators. The objective of the study is to design controls for robot manipulators without the knowledge of the boundary of the uncertainties/disturbances by using an adaptive sliding mode control while elucidating the robustness of the adaptive sliding mode control. A sliding mode control provides for ultimate accuracy in the presence of the bounded disturbances/uncertainties, although the sliding mode control also causes chattering. Chattering is undesirable for use with actual components, since it might cause damage to them with a subsequent loss of accuracy. Such chatter is caused by overestimation of the controller gain. An adaptive sliding mode is proposed as a solution to the problems created by chattering; to illustrate, a 2-link robot manipulator is simulated with an adaptive sliding mode control. The performance of each adaptive sliding mode control is demonstrated through the simulation results. The results of the simulations show the effectiveness for chattering mitigation by means of avoiding overestimation, and the robustness of an adaptive sliding mode control.