Informatica

An adaptive control scheme for mechanical manipulators is proposed. The control loop essentially consists of a network for learning the robot's inverse dynamics and on-line generating the control signal. Some simulation results are provided to evaluate the design. A supervisor is used to improve the performances of the system during the adaptation transients. The supervisor exerts two supervisory actions. The first one consists basically of updating the free-design adaptive controller parameters so that the value of a quadratic loss function is maintained sufficiently small. Such a function involves past tracking errors and their predictions both on appropriate time horizons of low performances during the adaptation transients. The supervisor exerts two supervisory actions. The second supervisory action consists basically of a on-line adjustment of the sampling period within an interval centered in a nominal value of the sampling period. The sampling period is selected so that the transient of the tracking error is improved according to the simple intuitive rule of using a sampling rate faster as the tracking error changes faster.

This paper addresses the application of convergence rules of gradient-type discrete algorithms to discrete adaptive control algorithms for linear time-invariant systems, which are based on Lyapunov's – like functions, in order to improve the transient performances based on fast adaptation. In particular, the adaptation covergence is increased as a generalized or filtered error increases through the application of Armijo rule for regulating the decrease of each Lyapunov's-like function on which the particular adaptation algorithm is based. The proposed scheme can be implemented with minor modifications in systems subject to unmodelled dynamics if some weak knowledge on such a dynamics is available consisting of upper-bounds of the dimension and norm of the unmodelled parameter vector.