VTOL Transition

Problem Statement

VTOL transition combines multirotor-style hover control with fixed-wing forward-flight aerodynamics. The key challenge is managing control authority handoff across flight regimes.

Model and Formulation

Blend hover and forward-flight controllers with scheduling variable \sigma \in [0,1]:

$$u=(1-\sigma )u_{hover}+\sigma u_{wing}$$

with \sigma scheduled by airspeed, pitch, and altitude envelopes.

Practical Notes

• A partial tilt is not a transition. Ramping the rotors to 30° never leaves rotor-borne flight: the wing is along for the ride and the demo is a slow quadrotor. The interesting regime is the handover itself, which only happens when the rotors go the full 90° and lift authority actually migrates to the wing. In cruise here the wing carries 95 % of the weight.

• Draw the tilt, don't imply it. This demo was previously rendered with a fixed-geometry quadrotor, which hid the one thing it exists to show: the rotors rotating through 90°. It now uses draw_vtol_3d, whose nacelles sit at the aircraft's actual tilt angle, so hover, transition and cruise are distinguishable from the picture alone.

• Let the mode machine own the schedule. VTOLController switches on measured airspeed against the wing's stall margin, so the tilt ramp is a consequence of the aircraft being ready rather than an open-loop timer that happens to work at one set of conditions.

• Watch altitude through the transition, not just at the ends. That is where a single altitude law has to degrade gracefully from pure rotor lift to pure wing lift; the atlas demo holds it to 4.8 m peak deviation.

• Transition corridors need explicit safety constraints.

• Inadequate gain scheduling causes pitch excursions and altitude loss.

• Propulsion and control-surface limits must be jointly managed.

Evidence

References

• Lustosa et al., Dynamic Transition for VTOL UAVs
• Sun et al., Review of Tilt-Rotor and VTOL Transition Control

Related Algorithms

• Fixed-Wing Flight
• PID Hover