Acoustic Signature Optimisation for Bioinspired Flapping-Wing Micro Aerial Vehicles

Yi Lun Duh, Krispin C. V. Broers, Sophie F. ArmaniniarXiv 2026 · arXiv

TL;DR

This paper presents a dual-strategy approach to quiet bioinspired flapping-wing micro aerial vehicles (FWMAVs) by optimizing wing geometry. By using a nonlinear structural model to optimize carbon-fiber venation patterns and adding bioinspired trailing-edge fringes, the researchers achieved up to a 7.8 dB reduction in impulsive "reversal clicks" and a 4 dB reduction in broadband "swoosh" noise without sacrificing lift.

What Problem Does This Paper Address?

While flapping-wing robots are often touted for their potential for quiet, unobtrusive flight, the reality is disappointing: engineered wings are typically 15–30 dB louder than the animals they mimic. This "acoustic gap" limits their use in sensitive environments like wildlife observation or indoor monitoring.

The authors identify two primary noise sources in membrane wings:

  1. The Reversal Click: An impulsive sound caused by the wing membrane "snapping through" between curvatures at the end of each stroke.
  2. The Aerodynamic Swoosh: Broadband noise generated by vortical flow shed at the trailing edge.

Previous research focused on performance (lift/thrust) rather than acoustics. This paper aims to fill that gap by treating wing venation (the stiffening ribs) and edge geometry as design variables for noise mitigation.

Methods

The researchers developed a comprehensive wing design framework that combines computational optimization with a novel manufacturing process.

1. Venation Optimization

To tackle the reversal click, the team modeled the wing as a von Kármán plate—a thin-sheet model that accounts for large out-of-plane deflections. Crucially, they identified that the click is a dynamic, inertia-driven event rather than a static buckling one. They implemented an iterative optimization loop (see Figure 3) using Bayesian optimization and a Gaussian process to find the quietest rib layout. The objective function was a "loading-rate surrogate"—essentially the high-passed rate of change of force the wing exerts on the air.

![Fig.

  1. An overview of the iterative process of the optimisation model.](/figures/art_4f1d2be89edc4d30bbe8917b83c21747)

2. Bioinspired Fringes

For the broadband swoosh, they looked to hummingbird feathers. They compared standard "cut" serrations (which remove wing area) against additive fringes—triangular teeth added beyond the mean trailing edge. This approach aims to disrupt vortex shedding without losing lifting surface.

3. Integrated Manufacturing

To ensure precision, the team moved away from gluing individual carbon rods. Instead, they used a CNC waterjet to cut single-piece wing frames (including the optimized venation) from cured carbon-fiber sheets, which were then bonded to a foam membrane.

Datasets

While this paper does not use a traditional large-scale dataset, it relies on:

  • Computational Evaluations: 180 unique wing designs were simulated during the Bayesian optimization process to map the design space.
  • Experimental Benchmarking: A custom flapping-mechanism test rig (Figure 2) was used to collect acoustic and lift data for a baseline wing (B0), two venation variants (M1, M1-adverse), and six edge-treatment variations (F0-F4 and a subtractive serration).

Key Insights and Ideas

  • Orientation Over Thickness: The optimization revealed that the angle of the venation ribs is far more critical for noise reduction than their thickness.
  • Diagonal Ribs are Key: The most effective designs placed ribs at a diagonal (roughly 48°) near the wing root. This is because the wing undergoes significant shear/twist during stroke reversal; a diagonal rib crosses these principal directions of twist, stiffening the membrane against the snap-through.
  • Additive vs. Subtractive: While cut serrations are common in airfoil research, additive fringes are superior for FWMAVs. They provide comparable noise reduction to cut serrations but generate significantly more lift (approx. 25% more) by maintaining the full membrane area.
  • Mass as a Self-Penalty: In the optimization, adding too much carbon fiber actually increased noise because the extra mass increased the inertial load at stroke reversal. The "quietest" wing was not the stiffest one, but the one with the best stiffness-to-mass distribution.

Key Results

The researchers successfully manufactured and tested several wing designs (Figure 4) to validate their models.

Manufactured wing models with the reference wing, the venation pattern set, and the edge-treatment set.
  • Venation: The optimized M1 design reduced the reversal click by 7.8 dB at the 90° microphone position compared to the baseline (Figure 5). Even a minor 5° change in rib angle (M1-adverse) almost entirely removed this benefit, highlighting the sensitivity of the design.
  • Fringes: The best fringe design (F2, short and coarse) reduced broadband noise by 4.1 dB (Figure 6).
  • Lift Performance: The optimized venation wing (M1) actually produced slightly more lift than the baseline (0.112 N vs 0.102 N).

![Fig.

  1. Reversal-averaged snap envelope at the 90◦ microphone position for baseline wing B0, venation set M1 and M1-adverse.](/figures/art_c1064ddaa6244e18b1e7f1a79eb89044)

Limitations and Future Work

The primary limitation noted was the mechanical noise floor. Above 1 kHz, the noise from the flapping mechanism's drive system dominated the acoustic signature, masking potential further improvements from the wing design.

Future research directions include:

  • Developing a quieter flapping mechanism to explore high-frequency noise reduction.
  • Testing a "combined" wing that features both optimized venation and additive fringes.
  • Investigating how these passive geometric changes perform across different flapping frequencies and amplitudes.

Summary updated

Abstract

Flapping-wing micro air vehicles are prized for quiet flight, yet the few existing measurements show engineered wings to be 15-30 dB louder than the animals they imitate. This paper investigates two passive geometric approaches to reduce the acoustic signature of a flapping membrane wing, and applies these to an exemplary wing. A significant sound source in flapping wings is the impulsive click radiated at stroke reversal, when a slack membrane snaps between two curvatures. Our first noise-reduction approach optimises the wing's venation layout, based on an analytical model of the reversal click as a dynamic snap-through of the membrane. The optimised venation pattern was manufactured and resulted in a reduction of up to 8 dB during the reversal snap in experiments. During testing, the venation was found to be sensitive to orientation rather than thickness. Secondly, an additive trailing-edge fringe inspired by hummingbird feathers was shown to further reduce the broadband swoosh by 3-4 dB at no lift cost, matching a cut serration wing without removing lifting area.
BibTeX
@misc{duh2026acoustic,
  author = {Yi Lun Duh and Krispin C. V. Broers and Sophie F. Armanini},
  title = {{Acoustic Signature Optimisation for Bioinspired Flapping-Wing Micro Aerial Vehicles}},
  year = {2026},
  eprint = {2610.37190},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.37190},
}
Figure