M. Heiss, INVERSE PASSIVE LEARNING OF AN INPUT OUTPUT-MAP THROUGH UPDATE-SPLINE-SMOOTHING, IEEE transactions on automatic control, 39(2), 1994, pp. 259-268
Citations number
74
Categorie Soggetti
Controlo Theory & Cybernetics","Robotics & Automatic Control","Engineering, Eletrical & Electronic
This paper presents a robust method of learning passively a one-dimens
ional input-output-map when receiving only indirect information about
the correct input-output-map (e.g., only the sign of the deviation bet
ween the actual estimated output value and the correct output value is
obtained). This information is obtained for only one input-output com
bination per updating cycle. The approach is to increment or decrement
step by step the output values of the actually stored map and then to
apply global or local cubic spline smoothing in order to avoid ''adap
tation holes'' at points which are never updated or less frequently up
dated than other points. This method works with noisy measurements as
well as slowly time-varying systems. Even continuous changes of the de
sired input-output-relation do not result in instability. Problems of
convergence and stability are treated and design rules are given.