Adaptive output recurrent neural network for overhead crane system

In this study, an adaptive output recurrent neural network (AORNN) controller is employed to control a practical overhead crane system with multi objective control problems. Trolley position error and swing angle error are used instead of a complex dynamic model to design the controller. The gradient descent method is adopted to adjust the AORNN parameters online. Moreover, an analytical method based on a Lyapunov function is proposed to determine the learning rates of the AORNN so that the convergence of the system can be guaranteed. Finally, the effectiveness of the proposed control system is verified by experiment and simulation of overhead crane system. The results show that AORNN control system can have a good performance in application.