Swashplateless-elevon Actuation for a Dual-rotor Tail-sitter VTOL UAV

Authors: Nan Chen, Fanze Kong, Haotian Li, Jiayuan Liu, Ziwei Ye, Wei Xu, Fangcheng Zhu, Ximin Lyu, Fu Zhang · IROS 2023

Licences

Open Access
Paper: CC BY 4.0

Everything below draws on the full paper under its CC BY 4.0 licence; its figure is credited to the authors.

TL;DR

Pitch moves from the elevons to passive swashplateless rotor hubs driven by motor speed modulation: a 2.25 kg dual-rotor tail-sitter takes off, tracks and rejects wind better, with no added actuators

Why it matters

With pitch handled by the rotors, elevon travel stays available for yaw and disturbance rejection, and the pitch moment does not depend on propeller airflow over the elevons, so take-off near the ground stays stable. This is achieved without extra motors or servos, keeping the dual-rotor layout that the paper notes is cheaper, lighter and more portable than quad-rotor tail-sitters.

Problem

Dual-rotor tail-sitter VTOL UAVs usually control both pitch and yaw with elevons (conventional elevon actuation, CEA), and the large wing catches wind mainly in those two axes. Large control demands in both directions saturate the limited elevon deflection, degrading control or destabilizing the vehicle. Near the ground the propeller downwash is disturbed, so the elevons lose control moment during take-off and pitch can become unstable.

Proposed solution

Swashplateless-elevon actuation (SEA) mounts a passive swashplateless mechanism on each motor: sinusoidal speed modulation of the motor, timed by a magnetic encoder reading of rotor angle, makes asymmetrically hinged blades change pitch cyclically and produces a pitch moment. Roll comes from differential thrust, yaw from the elevons alone, and a custom mixer added to PX4 runs at 1 kHz to map desired thrust and moments to motor throttles and servo angles. The improved hub adds ball bearings and pressure bearings to cut hinge friction.

Key insights

A passive swashplateless hub produces a pitch moment from motor speed modulation alone, so pitch control needs no added actuator.
Sending pitch to the rotor hubs and yaw to the elevons stops the elevons from saturating when the wing takes large disturbances in both axes.
The swashplateless moment is independent of propeller airflow over the control surfaces, so ground effect at take-off does not degrade it.
The SEA mixer runs at the IMU rate, set to 1 kHz so that 910 Hz encoder readings are not dropped, and the DShot600 ESC protocol's 26.7 µs delay is short enough for high-frequency speed modulation.
Ball and pressure bearings in the hinges give a smoother, more linear moment output and faster response than the original swashplateless design.

Main results

Take-off under position control: position error 0.04 m with SEA vs 0.79 m with CEA (5.1%), with CEA pitch error reaching 14.7°.
Take-off under attitude control: maximum pitch error 1.4° with SEA, against 8.6° for CEA from the ground and 2.1° for CEA from a raised pedestal.
Figure-of-eight tracking (2 m × 1 m, continuous yaw): median position error 14.17 cm vs 15.35 cm (max 38.91 vs 40.64 cm) and median yaw error 19.5° vs 27.1° (max 87.9° vs 119.3°), SEA vs CEA.
Hover in a balanced fan gust of about 4.5 m/s: median x-position error 1.77 cm vs 3.83 cm (max 6.60 vs 7.92 cm); CEA had a slightly smaller median pitch error (1.09° vs 1.32°) but a larger maximum (5.78° vs 4.08°).
Position steps under one-sided wind: in the forward step CEA showed pitch oscillation, yaw error up to 31.2° and roll error up to 8.2°, with no oscillation for SEA; in the lateral step maximum yaw error was 22.1° for CEA vs 8.4° for SEA.
Outdoor transition and fixed-wing flight with SEA: transition tracked without overshoot, speed reached 9.6 m/s, and maximum errors after transition were 5.3° in roll and 3.8° in yaw. Section IV-D gives the fixed-wing pitch setpoint as -65° (wing at 35° to horizontal) and then says pitch 'always remained at 65°'.

Limitations

SEA was not compared quantitatively with CEA in fixed-wing flight; the authors cite earlier verification of CEA fixed-wing flight and only show SEA feasibility.
In fixed-wing flight, roll (motors) and yaw (elevons) showed slight vibrations around the setpoints, largest just after transition, which the authors say may come from airflow disturbance.
Under balanced wind in hover, CEA had a slightly smaller median pitch error than SEA (1.09° vs 1.32°).
The blade-inertia delay constant γ₀ must be calibrated in advance from test-stand data.
All results come from one custom airframe (Hong He), with indoor position feedback from a Vicon motion-capture system; the paper has no separate limitations section.

Field impact

Winner of the IROS 2023 Best Paper Award on Robot Mechanisms and Design (sponsored by ROBOTIS) and a finalist for IROS 2023 Best Overall Paper and Best Student Paper; cited 12 times. The authors publish a video of all experiments; the paper gives no code or design-file release.

Abstract

In this paper, we propose a novel swashplateless-elevon actuation (SEA) for dual-rotor tail-sitter vertical takeoff and landing (VTOL) unmanned aerial vehicles (UAVs). In contrast to the conventional elevon actuation (CEA) which controls both pitch and yaw using elevons, the SEA adopts swashplateless mechanisms to generate an extra moment through motor speed modulation to control pitch and uses elevons solely for controlling yaw, without requiring additional actuators. This decoupled control strategy mitigates the saturation of elevons' deflection needed for large pitch and yaw control actions, thus improving the UAV's control performance on trajectory tracking and disturbance rejection performance in the presence of large external disturbances. Furthermore, the SEA overcomes the actuation degradation issues experienced by the CEA when the UAV is in close proximity to the ground, leading to a smoother and more stable take-off process. We validate and compare the performances of the SEA and the CEA in various real-world flight conditions, including take-off, trajectory tracking, and hover flight and position steps under external disturbance. Experimental results demonstrate that the SEA has better performances than the CEA. Moreover, we verify the SEA's feasibility in the attitude transition process and fixed-wing-mode flight of the VTOL UAV. The results indicate that the SEA can accurately control pitch in the presence of high-speed incoming airflow and maintain a stable attitude during fixed-wing mode flight. Video of all experiments can be found in youtube.com/watch?v=Sx9Rk4Zf7sQ

Open Source Robotics