A learning system possesses the capability to improve its performance
over time by interaction with its environment. A learning control syst
em is designed so that its learning controller has the ability to impr
ove the performance of the closed-loop system by generating command in
puts to the plant and utilizing feedback information from the plant. I
n this brief article, we introduce a learning controller that is devel
oped by synthesizing several basic ideas from fuzzy set and control th
eory, self-organizing control, and conventional adaptive control. We u
tilize a learning mechanism that observes the plant outputs and adjust
s the membership functions of the rules in a direct fuzzy controller s
o that the overall system behaves like reference model. The effectiven
ess of this fuzzy model reference learning controller is illustrated b
y showing that it can achieve high performance learning control for a
nonlinear time-varying rocket velocity control problem and a multiinpu
t multi-output two-degree-of-freedom robot manipulator. (C) 1996 John
Wiley and Sons, Inc.