Neural network controllers for a flexible-link manipulator: experimental results

In this paper, the problem of tip position tracking of a flexible-link manipulator is considered. Four different neural network schemes are presented. In the first scheme, the controller is designed based on tracking the hub position while controlling the elastic deflection at the tip. In the second scheme, an output redefinition through online learning is proposed. To improve the transient as well as steady-state response of the system the two schemes are modified by adding a joint PD controller to the neural network controller. The performance of the proposed neural network controllers is illustrated by experimental results on a single flexible-link testbed. The control structures developed assume no a priori knowledge about the system dynamics, and the networks are all trained and employed as online controllers and no off-line training is required.