Disturbance Observer based Fractional Order Iterative Learning Control of Helicopter Model

Abstract

The Twin Rotor Aerodynamic System (TRAS) provides a general representation of the aerodynamic characteristics of helicopters as well as other hovering rotor vehicles. Due to the nonlinear nature of the system and the substantial cross-coupling between the inputs and outputs of the main and tail rotors, managing such a system for either stabilization or reference tracking is a challenging challenge. The reference tracking and disturbance rejection problem for a Multi-Input Multi-Output (MIMO) TRAS is investigated in this study using a hybrid architecture based on a Fractional Order Proportional Integral Derivative (FOPID) controller with an Extended Kalman Filter (EKF) as a Disturbance Observer (DOB) and Proportional Derivative Iterative Learning Control (PD-ILC) (FOPIDEKF-DOB-PD-ILC). The system is divided into subsystems for the main and tail rotors. Since TRAS is unstable and ILC is only used for stable systems, FOPID controller is employed to stabilize the plant. Sequential quadratic programming (SQP) is used to calculate the parameters of the FOPID controller. To lessen tracking error and gradually enhance system performance, PD-ILC is applied as a feedforward controller. ILC is employed in this situation as an external controller with no effect on the current control architecture. The FOPID-EKF-DOB controller adjusts the new control input according to the PD-ILC control law. Results from the simulation are used to evaluate the efficiency of the suggested strategy based on performance metrics.