Investigating the integration of biosonar sensing and flight control in bats on Borneo

Abstract

Many bat species rely on biosonar as their primary source of sensory information about their environments. Species that are able to navigate amid dense vegetation use this biosonar information to guide highly dexterous flight maneuvers. Understanding the connection between biosonar inputs and flight outputs poses a challenge, because the flight apparatus of bats has the most degrees of freedom of any flight system—whether natural or man-made. At the same time, the biosonar inputs consist of just two one-dimensional acoustic time signals (i.e., echoes received at the two ears). Due to the complexity and large variability in the echoes and the flight maneuvers of bats, understanding the input-output relationships requires the ability to collect large amounts of quantitative data on the acoustics and the flight kinematics. In addition, comparing different bat species could offer a window into the principles behind the evolutionary co-adaptation of biosonar and flight. To accomplish this, a cylindrical flight tunnel that integrates synchronized arrays of 50 high-speed video cameras and 32 ultrasonic microphones has been set up on the island of Borneo. This instrument is complemented by a set of custom deep-learning techniques that can handle the large amounts of data that are being produced.