On the dynamics of a neural network for robot trajectory tracking

In this paper, the dynamic behavior of a three-layer feedforward neural network as a uncertainty compensator for robotic control is investigated. The investigation is conducted in the context of the robot trajectory tracking problem, where the neural network (with the errorbackpropagation algorithm) is used as a uncertainty compensator in conjunction with the feedback linearizat ion control (i.e. computed torque) and a PD control. ugh computer simulation, it is verified that the dynamics of the neural netwo s a specific pattern when learning rate is sufficiently small, and that such a specific pattern of weight variation in the neural network represents a suficient condition for closed-loop system performance improvement.