Machine learning reveals the control mechanics of the insect wing hinge

Abstract

Insects constitute the most species-rich radiation of metazoa, a success due primarily to the evolution of active flight. Unlike pterosaurs, birds, and bats, the wings of insects did not evolve from legs, but are novel structures attached to the body via a biomechanically complex hinge that transforms the tiny, high-frequency oscillations of specialized power muscles into the sweeping back-and-forth motion of the wings. Due to the minute size and morphological complexity, the basic mechanics of the hinge are poorly understood. The hinge consists of a series of tiny, hardened structures called sclerites that are interconnected to one another via flexible joints and regulated by the activity of a set of specialized steering muscles. In this study, we imaged the activity of these steering muscles in a fly using a genetically encoded calcium indicator, while simultaneously tracking the 3D motion of the wings with high-speed cameras. Using machine learning approaches, we created a convolutional neural network that accurately predicts wing motion from the activity of the steering muscles, and an autoencoder that predicts the mechanical role of the individual sclerites on wing motion. By replaying patterns of wing motion on a dynamically scaled robotic fly, we quantified the effects of steering muscle activity on the production of aerodynamic forces. A physics-based simulation that incorporates our model of the wing hinge generates flight maneuvers that are remarkably similar to those of free flying flies. This integrative, multi-disciplinary approach reveals the mechanical control logic of the insect wing hinge, arguably the most sophisticated and evolutionarily important skeletal structure in the natural world.