Advisor: 

Dr. Simon Sponberg (Georgia Institute of Technology) 

Committee: 

Dr. Saad Bhamla (Georgia Institute of Technology) 

Dr. Nick Gravish (University of California, San Diego) 

Dr. David Hu (Georgia Institute of Technology) 

Dr. Kurt Wiesenfeld (Georgia Institute of Technology) 

Beyond resonance: synchronous and stretch-activated actuation in insect flight 

The generation of high power, rhythmic movement is a common feature of biological and robotic locomotion. Insects stand out among these systems because wingbeat frequencies are often an order of magnitude greater (up to 1000 Hz) and face the extreme energetic costs of flapping wing flight. Given the oscillatory nature of insect flight, insects are believed to be resonant. In this framework, elastic structures significantly reduce inertial power costs by storing and returning excess kinetic energy during a wing stroke. However, evidence suggests that a resonance model of flight is incomplete. Unlike the time-periodic (e.g. sinusoidal) forcing of resonant systems, many insects have evolved strain-dependent muscles. Pairs of these muscles excite each other independently from neural input.  

This thesis explores how strain-dependent actuators coupled to deformable systems generate high power, rhythmic movements. In Chapter 1, we identified how spring-like properties emerge from heterogeneous exoskeletal shell. Notably, the exoskeleton alone satisfies the energy exchange demands of flight. In Chapter 2, we perturbed hawkmoths and discovered the capacity for +/- 16% frequency modulation at the wingstroke timescale. Unlike their robotic counterparts that explicitly abdicate frequency modulation in favor of energy efficiency, frequency modulation is an underappreciated control strategy in insect flight. In Chapter 3, we developed a mechanical model of hawkmoth mechanics and found that wingbeat frequencies are 50% above resonance. These results suggest that resonance tuning is neither a ubiquitous nor necessary feature of insect flight. Finally, in Chapter 4, we introduced both time periodic and stretch activated forcing to the passive mechanical system. We discovered that a small set of parameters drive transitions between synchronous and self-excited wingbeats. This single dynamical system explains evolutionary transitions in insects and generalizes to robotic systems.