Neural network ARMAX model for a Furuta pendulum

Abstract

The rotational inverted pendulum or Furuta Pendulum is a mechatronic system used by control engineers to explore various dynamic modeling and control schemes. Due to its nonlinear nature, open-loop instability, and under-actuated system, it serves as a basis for designing vehicles like the Segway, self-balancing scooters, and hoverboards. The authors present a model for the Furuta Pendulum using Euler-Lagrange equations and a methodology for identifying a black-box model by training a NNARMAX (Neural Network Auto-Regressive Moving Average with exogenous inputs). The results show that two interconnected MISO-NNARMAX models accurately estimate 10-step-ahead predictions for the horizontal and vertical angles.