Experimental experience on iterative learning control implemented on a prototypal manipulator

The implementation of iterative learning control algorithms on a real mechanical manipulator is discussed. The algorithms rely on off-line estimates of generalized velocity and acceleration errors, and on their boundedness in presence of noisy measures. A class of robust estimators is introduced, based on the theory of the spectral analysis of descriptor systems. The use of these estimators in the implementation of learning control algorithms has been tested on a prototypal manipulator at DIST and experimental results are reported.<>