This study investigates the nonlinear dynamics and control strategies for a Cessna-182 unmanned aerial vehicle (UAV). The longitudinal and lateral dynamic models were derived using aerodynamic analysis conducted in ANSYS Fluent. Various control methodologies, including PID (proportional, integral, derivative) tuning with genetic algorithm (GA), root locus (RL), MATLAB tuned (MT), Ziegler Nichols (ZN), the model reference adaptive controller (MRAC), linear quadratic regulator (LQR) and linear quadratic Gaussian (LQG), were implemented and compared through simulations. Due to the inadequacy of PID control under variable environmental and noise conditions, LQR and LQG controllers, including the full state-space model required for real-time applications, are simulated for both longitudinal and lateral motions in addition to the PID controller. Although the LQR controller has acceptable simulation results for noiseless conditions, the superiority and stable structure of the LQG controller under noise and disturbance effects are highlighted. The results, including noise Dryden turbulence effects, highlight the advantages of proposed MRAC and LQG for robust stability and precise flight performance. Experimental flight tests validate the theoretical findings, demonstrating the practical viability of the proposed control approaches. The results obtained for the Cessna-182 mini-UAV will be effective for future researchers in terms of obtaining UAV dynamics and also evaluating different control strategies of the UAV.

This paper reports on a new navigation algorithm for fixed-wing Unmanned Aerial Vehicles (UAVs) to bridge Global Position System (GPS) outages, based on a common navigation system configuration. The ground velocity is obtained from wind-compensated airspeed, and a centripetal force model is introduced to estimate the motion acceleration. Compensated by this acceleration, the gravity vector can be extracted from the accelerometer measurement. Finally, fusing the information of the ground velocity, magnetic heading, barometric height, and gravity vector, the Integrated Navigation System (INS) is reconstructed, and an Extended Kalman Filter (EKF) is used to estimate INS errors. Hardware-in-loop simulation results show that compared with INS-only solutions, the proposed method effectively resists long-term drift of INS errors and significantly improves the accuracy for dynamic navigation during GPS outages.