Robot manipulator control based on the inner-learning mechanism

This paper deals with the control problem of a class of multi-tasking and varying-load robot systems, presents an effective control architecture based on inner-learning mechanism. The novelty of the controller designed by the new architecture is in the fact that the sub-controllers can learn from each other and change information from each other; it can also avoid the failures of the sub-controllers. The good performance and the feasibility of this approach are shown through the experimental results.