Revealing Principles of Adaptive Flight Control in Insects using a Virtual Reality Flight Simulator
Open Access
- Author:
- Jaworski, Eric
- Area of Honors:
- Mechanical Engineering
- Degree:
- Bachelor of Science
- Document Type:
- Thesis
- Thesis Supervisors:
- Jean-Michel Mongeau, Thesis Supervisor
Daniel Humberto Cortes Correales, Thesis Honors Advisor - Keywords:
- Flight
Adaptive Control
Drosophila - Abstract:
- Flying insects are highly agile and capable of performing complex aerial maneuvers that are unmatched by current flying robots. Among flying insects, the fruit fly Drosophila melanogaster is a highly agile and adaptable flier. It has a complex control system that allows it to react to a host of environmental perturbations. Flies must be able to sustain flight under different loading conditions (e.g. following an increase in mass), but their ability to fly with different levels of inertia is poorly understood. The goal of this study was to analyze the effect of adding inertia about the yaw direction on the ability of fruit flies to stabilize their gaze and maintain adequate flight performance. With increased interest in developing micro-scale flying robots, flying insects can provide inspiration to develop more fault-tolerant robots. Using a virtual reality flight simulator, fruit flies were glued to a pin fixed between two magnets, restricting its movement to rotation about the yaw axis. To add inertia to the fly, hollow cylinders with 1x, 2x, 4x, 8x, 16x, 32x, and 64x the inertia of a fruit fly were designed, 3D printed, and then glued to the pin. In the arena, flies were subjected to a complex sum-of-sines visual stimulus and their body angular heading was measured. The response of the flies was analyzed in the frequency domain, and the gain and phase of the response at each inertia level was measured. First-order transfer functions were fit to the frequency domain response of the flies, and it was discovered that neither the open-loop gain nor pole location of the transfer function decreased proportionally with the added inertia. Furthermore, the time constant of the first-order system remained fairly constant between levels of inertia. These results show that flies can actively modulate damping when subject to different levels of inertia, thus maintaining stability. Although the exact mechanism to adjust damping remains unknown, this study provides insight into a potential method that flying robots could use to account for changes in inertia during flight and to improve their stability.