Session: 4.1.1 - Interactions in Bio-Inspired Propulsion

Paper Number: 158136

158136 - Kinematic Characterization of Bee Flight Modes Using Deep Learning for Prospective Fluid Flow Analysis 

Abstract: 

Kinematic Characterization of Bee Flight Modes using Deep Learning for Prospective Fluid Flow Analysis

Clayton F. Fernandes¹, Chi Nnoka¹ and Javid Bayandor¹

Affiliations:   

¹CRashworthiness for Aerospace Structures and Hybrids (CRASH) Lab
 Department of Mechanical and Aerospace Engineering
 University at Buffalo - The State University of New York

This study investigates the wing kinematics and fluid dynamics observed in different flight modes of the Western Honey Bee (Apis mellifera) including take-off, hovering, landing and fanning using a combination of high-speed video capture and machine learning. Wing motion is first recorded in both free-flight and controlled laboratory conditions using a high-speed camera system. Using pose estimation-based machine learning, the wing trajectories from different views are recreated. These wing trajectories provide inputs for future computational fluid dynamics (CFD) simulations for analyzing vortex formation and lift-generation mechanisms during the different flapping modes.

Different types of flow patterns and aerodynamic characteristics may be observed across different flight modes. During take-off, bees utilize rapid high-amplitude wingbeats to generate the necessary lift and thrust which along with leading-edge vortices enable initial lift generation. Hovering wing flapping shows vortex shedding and potential wake recapture mechanisms involved in energy-efficient lift generation. During landing, wing kinematics prioritize stability, lift reduction and aerodynamic braking. The fanning mode, which is used for thermoregulation for both the individual bee and the hive when swarming, consists of wing motion that generates directional airflow with low energy expenditure.

This study aims to highlight any significant trade-offs between stability, maneuverability and energy efficiency in bee flight. Integrating experimental data with numerical modeling helps in studying the aerodynamic flow of flapping wing insects and furthering understanding of fluids engineering. The research focuses on using wing kinematics as input for future CFD simulations which may help further elaborate the mechanisms of vortex interactions, wake capture and aerodynamic efficiency, offering insight into how bees achieve flight and how they can be adapted into multifunctional bioinspired or biomimetic designs with potential applications that include micro-air vehicles for search and rescue missions, precision agriculture and environmental monitoring.

Presenting Author: Clayton F. Fernandes University at Buffalo - The State University of New York

Presenting Author Biography: Clayton F. Fernandes is a PhD. Candidate in Mechanical Engineering at the CRashworthiness of Aerospace Structures and Hybrids (CRASH) Lab in the Department of Mechanical and Aerospace Engineering at the University at Buffalo. His work specializes in bioinspired robotics, morphing systems and data acquisition with a focus on novel propulsion systems for flapping wing air vehicles.