Practical aspects on robust control strategies of a single link flexible manipulator

An investigation is conducted of an adaptive approach to the control of a flexible robotic manipulator in the presence of changing payload. The approach is a combination of indirect adaptive control and gain scheduling. A lookup table of precalculated controller parameter vectors for a set of possible payloads is utilized. The payload in the lookup table closest to the actual payload provides the initial controller parameter vector that sets the starting adaptation point, and online recursive-least-squares identification of the plant parameters, together with online pole placement controller design, allows fast tuning of the adaptive controller. Simulation experiments, using realistic data obtained from the CSU experimental flexible manipulator testbed, show the feasibility of the approach