THE ROLE OF ADAPTIVE CONTROL SYSTEMS IN ENSURING FLIGHT STABILITY OF QUADCOPTERS

Abstract

This paper examines the challenge of maintaining stable flight for quadcopters under varying physical and atmospheric conditions. Traditional PID and LQR control algorithms often fail to cope in real-time with disturbances such as wind pressure changes, variations in air density, and payload shifts. To address these limitations, we propose an adaptive-robust control architecture based on the integration of Sliding Mode Control (SMC) and Radial Basis Function neural networks (RBF-NN). Our approach continuously gathers data from multiple sensors (IMU, barometer, anemometer, GPS) and employs state-estimation algorithms (Kalman filter, sigma-filtering) to identify the system’s dynamics online. Control laws are then optimized in real time to adjust to evolving conditions. Experimental results demonstrate that the SMC-RBF integration maintains trajectory error within 0.15 m RMS, reduces energy consumption by 10–12 %, and restores stability within 0.3 s after a disturbance. These findings confirm that adaptive control systems significantly enhance the safety, reliability, and endurance of quadcopters, laying the groundwork for the next generation of innovative aviation technologies.