Multi-Objective PID Controller Optimization for Quanser Aero 2 Using NSGA-II: Simulation and Hardware Implementation

Abstract

This paper addresses the challenge of optimal Proportional-Integral-Derivative (PID) controller tuning for coupled Multi-Input Multi-Output (MIMO) mechatronic systems and validates the solution on a Quanser Aero 2 dual-rotor helicopter platform. The primary challenge in controlling such systems lies in
satisfying conflicting performance objectives, namely, minimizing tracking error while simultaneously reducing control effort to ensure energy efficiency and respect physical actuator limitations. To address this, the controller tuning problem is formulated as a dual-objective optimization problem. The Non-dominated Sorting Genetic Algorithm II (NSGA-II) is applied to obtain a diverse set of Pareto-optimal solutions that characterize the inherent compromise between tracking accuracy and control effort. The methodology is validated through simulation and experimental validation, demonstrating significant performance gains over a conventional PD controller. A rigorous comparative analysis using multiple performance indices (ISE, ITAE, RMSE) provides a quantitative assessment of the different solutions on the Pareto front. The results confirm that the NSGA-II approach provides a systematic and effective method for designing high-performance, robust, and practical controllers for complex MIMO systems. The compromise solution from the Pareto front offers an excellent balance of responsiveness, stability, and energy efficiency in both simulation and real-world experiments.