1994 Volume 7 Issue 4 Pages 134-141
In this paper, the adaptive multi-step learning control for continuous-time linear system is studied. Input function uk+1 (t) is calculated from the triplets (uk, ek, ek) and (uk-1, ek-1, ek-1) where ek is tracking error of k-th trial. The weighting coefficients of the triplets are determined adaptively. The convergence of the control law is proved by means of evaluation of the norm of tracking error. Moreover, the learning control is applied to the tracking problem of two-link robot manipulator.