An approach to adaptive control of robotic manipulators (Q1071709)

The paper considers the problem of the decentralized control synthesis for large-scale mechanical systems with uncertain and variable parameters. In particular, the control synthesis for robotic systems with partially unknown parameters is discussed. A manipulator system is viewed as a set of subsystems, each associated to one degree of freedom of the mechanism, and for each subsystem a local stabilizing controller is synthesized. In result, a robust non-adaptive decentralized control algorithm is obtained, stabilizing robots for all allowable variations of the parameters. Another two-stage (indirect) adaptive control algorithm, where the estimation of the robot parameters is performed first and then used for adjusting the coefficients of local regulators, is proposed for the case when the robust non-adaptive control cannot withstand all expected variations of parameters. Both approaches are illustrated by simulation of an industrial robot with unknown payload mass, and some numerical experiences are presented.