Authors : Mustefa Jibril, Messay Tadese and Nuriye Hassen
Abstract: In this study, the performance of inverted pendulum has been Investigated using neural network control theory. The proposed controllers used in this study are NARMA-L2 with resilient backpropagation and Levenberg Marquardt backpropagation algorithm controllers. The mathematical model of Inverted Pendulum on a Cart driving mechanism have been done successfully. Comparison of an inverted pendulum with NARMA-L2 with resilient backpropagation and Levenberg Marquardt backpropagation algorithm controllers for a control target deviation of an angle from vertical of the inverted pendulum using two input signals (step and random). The simulation result shows that the inverted pendulum with NARMA-L2 with resilient backpropagation controller to have a small rise time, settling time and percentage overshoot in the step response and having a good response in the random response too. Finally, the inverted pendulum with with NARMA-L2 with resilient backpropagation controller shows the best performance in the overall simulation result.
Mustefa Jibril, Messay Tadese and Nuriye Hassen, 2021. Inverted Pendulum Control using NARMA-L2 with Resilient Backpropagation and Levenberg Marquardt Backpropagation Training Algorithm. Journal of Engineering and Applied Sciences, 16: 324-330.