Dynamic Mode Decomposition of Wing Kinematics and Aerodynamic Force Production in Straight Flight of the Great Roundleaf Bat (Hipposideros armiger)

This study applies Dynamic Mode Decomposition (DMD) to motion-capture-based kinematics of straight flight in the great roundleaf bat (Hipposideros armiger) to identify dominant frequency components and evaluate their aerodynamic effects using prescribed-kinematics CFD. Three complex-conjugate mode pairs capture the primary deformation content: a zero-frequency static mode representing mean posture, a fundamental flapping mode at 9.2 Hz, and a superharmonic twisting mode at 18.6 Hz. Spatial localization shows the flapping mode represents the global stroke motion, with amplitude increasing toward the wingtip, whereas twisting is concentrated on the outboard wing, strongest near the distal trailing edge. Two temporal descriptions are compared: eigenvalue-based predictive dynamics with nearly constant-amplitude sinusoidal evolution, and data-projected dynamics exhibiting stroke-phase-dependent amplitude modulation. Aerodynamic results show that reconstructions retaining the static and flapping modes reproduce the dominant downstroke lift trend but underpredict thrust, particularly near stroke transitions. Incorporating the twisting mode substantially improves thrust recovery and increases lift recovery by preserving distal wing reorientation and associated pressure-difference signatures. Predictive dynamics provide actuator-ready commands but can introduce phase offsets relative to the measured, phase-modulated kinematics. Overall, bat straight flight kinematics support a compact frequency basis, but realistic force production requires stroke-phase-dependent modulation rather than purely harmonic evolution.