Do iridescent butterflies actively orient themselves to optimize their color signals?

Methods

Summary

I will be using the Raspberry pi (a small form Linux based computer system) as a basis for the data acquisition. The Raspberry pi system will use three main components: a camera, an Inertial Measurement Unit (IMU), and a GPS receiver. All these systems will be programmable and integrable through Linux using Python. The camera system can be programed to start video within a millisecond of the Raspberry pi internal clock. It will take video up to 90 fps progressively scanned. Millisecond timing will be satisfactory for this stage of the project giving only 0.5 cm location jitter for an insect flying at 5 m/s. The IMU will give the angular orientation of the camera. The GPS module will give a timing pulse that will be in-sync with the absolute GPS time to an accuracy much finer than the millisecond jitter associated with the video acquisition. The GPS location will have an error of at least 1.0 m so the distance between the cameras will have to be fixed and measured throughout the measurement. A finer scale GPS system is being investigated and may be available to the project within the budget. I will develop extensive Python code that will integrate these environmental data in a 3D framework. I will investigate the possibility of the cameras communicating together on a local computer to computer Wi-Fi network to send location and orientation information together to aid in coordinated visual tracking of flying insects.

Challenges

At each point in the integration of the system components runs the risk of incompatibility, or requiring more tools to make the system compatible and integrable. As in times past, collaboration and research have been key ingredients for solving compatibility issues.

Pre Analysis Plan

The system development will have one key operating principle—develop the system so that the analysis will be as straight forward as possible. To accomplish this, the system will be developed so that calibration must be performed before data can be acquired. Data will be stored and labeled based on the absolute GPS time and location. Videos of flying insects will be hand tracked using Python interface programs, which can potentially be made into a web based crowd sourcing analysis project—a citizen scientist based tracking program of flying insects. The geometries of the flight paths will ultimately be broken down to angular based statistics.