Optimal Second Order Sliding Control for the Robust Tracking of a 2-Degree-of-Freedom Helicopter System based on Metaheuristics and Artificial Neural Networks

Abstract

This paper presents a novel formulation for a Second-order Sliding Controller (SOSMC) for the trajectory tracking of a 2-degree-of-freedom (DoF) helicopter system based on the combination of a robust Second-order Discrete Sliding Mode Controller (SDOSMC) and a Linear Quadratic Regulator (LQR). The combination of these two approaches guarantees the effectiveness of the SOSMC in the face of uncertainties and disturbances. A new approach based on the resolution of the Sylvester equation is proposed in order to design a sliding surface that would ensure the optimal performance of the Sliding Mode Control (SMC). The performance of the SMC heavily depends on the selection of the sliding surface. The reformulation of the control problem is treated as a multi-objective optimization problem, and a new objective function based on dynamic aggregation is proposed for this purpose. The main contribution of this article lies in using the set of solutions provided by metaheuristics to leverage the capabilities of Deep Learning (DL) and to predict the optimal solution in terms of performance. The proposed control methodology is applied to control the pitch and yaw axes of the Quanser helicopter system. The simulation results, along with a comparative analysis were included to support the importance and efficacy of the suggested control technique.