Robust Control of Robotic Manipulators in the Presence of Dynamic Parameter Uncertainty

摘要

Sufficient conditions are proved for a robotic manipulator controller so that asymptotic tracking/regulation occurs, independent of dynamic parameter uncertainty, for a certain class of input signals. The uncertainty can be quite large, and arise chiefly from the manipulation of payloads with unknown mass/inertia properties. The control is obtained using a robust controller which consists of two separate parts: 1) a compensator which makes the closed-loop robotic system insensitive to parameter uncertainty and generates asymptotic regulation of a certain class of input signals and 2) a stabilizing compensator, whose purpose is to stabilize the closed-loop system. Stability of the closed-loop system is guaranteed by choosing large feedback gains. In addition to the above, it is also shown that the proposed feedback controller provides an arbitrarily small tracking error capability for the particular class of input trajectories.