STABILITY ANALYSIS AND PID CONTROL OPTIMIZATION OF A HYBRID VTOL UAV DURING HOVER-TO-FORWARD FLIGHT TRANSITION USING MATLAB/SIMULINK

Hybrid vertical take-off and landing (VTOL) unmanned aerial vehicles combine the hovering capability of multi-rotors with the endurance of fixed-wing aircraft, but the hover-to-forward-flight transition is the most demanding phase of the mission: the dominant source of lift migrates from the lift rotors to the wing while the pitch dynamics change from marginally unstable to aerodynamically stabilized. This paper presents a stability analysis and a PID control optimization for a quad-plane hybrid VTOL UAV during this transition. A nonlinear longitudinal three-degree-of-freedom model with rotor drag, rotor flapping moment, stall-blended aerodynamics and actuator dynamics is developed in a form that maps directly onto a MATLAB/Simulink block diagram. The model is linearized at six trim airspeeds between 0 and 22 m/s, showing an unstable oscillatory pair at hover (σ = +0.80 s⁻¹, ω = 1.41 rad/s) and an unstable real pole of +0.99 s⁻¹ at 5 m/s, which disappear as dynamic pressure grows. A gain-scheduled cascade of three PID loops (attitude, altitude and airspeed) with airspeed-based control allocation between lift rotors and elevator is then tuned by particle swarm optimization (PSO) using a cost that combines altitude and attitude ITAE, airspeed tracking, control effort and a phase-margin constraint of 45°. Compared with a manually tuned baseline, the optimized controller reduces the peak altitude deviation during transition from 8.0 m to 3.7 m, the altitude ITAE by 6.3 times, and the airspeed overshoot from 5.8% to 2.6%, while keeping the pitch-loop phase margin at or above 45° at every airspeed. In a 40-run Monte Carlo study with ±15% mass, reduced static margin, inertia and drag uncertainty and a 3 m/s vertical gust, all runs were stable for both controllers, and the optimized controller kept the worst-case altitude deviation to 8.6 m against 20.7 m for the baseline.