Science Literature

Anyone attempting to swat a fly will become aware of its remarkable aerodynamic capabilities. Its speed of response and ability to change direction abruptly far exceed our own powers as pursuers. The flight of insects has received considerable attention from researchers and some recent work was stimulated by the recognition of a gap in knowledge. The scientists realized that the previously-studied flight control system involving vision cannot be the explanation for how flies maintain stability in the face of unpredictable short disturbances.

"Corrective behavior often takes advantage of vision. For fruit flies, however, reaction time to visual stimuli is at least 10 wingbeats, so these insects must employ faster sensory circuits to recover from short time-scale disturbances and instabilities."

To study how fruit flies recover from in-flight disturbances, researchers glued magnetic pins to the insects' backs and zapped them with a magnetic field. This fly has a 1.5 mm pin on its back and is held in place by the tip of a sewing needle. (Credit: Wang, Cohen and Guckenheimer labs, Source here)

The experimental work required the research team to abandon ideas of tethering insects or imposing other restrictions on flight behavior. They needed to observe insects in free-flight.

"To probe this fast control strategy, we devised an experimental method that imposes impulsive mechanical disturbances to flying insects while allowing us to measure relevant aspects of flight behavior. We first glue tiny ferromagnetic pins to fruit flies and image their free flight using three orthogonally oriented high-speed video cameras. When a fly enters the filming volume, an optical trigger detects the insect, initiates recording, and activates a pair of Helmholtz coils that produce a magnetic field. The field and pin are both oriented horizontally, so the resulting torque on the pin reorients the yaw, or heading angle, of the insect. We then use a new motion tracking technique to extract the three-dimensional body and wing motions."

What they observed is that prior to the perturbation (which lasted 5ms, or about one wingbeat period), the wings beat symmetrically. After the magnetic torque was applied, 3 wingbeats were needed for the control system to respond, and then "asymmetries in the wing motions appear for about five wingbeats, indicating the insect is actively generating corrective torque". For small perturbations, the insects correct "nearly perfectly", whereas larger perturbations - although corrected to some extent - lead to permanent changes in heading.

"The accuracy of the recovery indicates that a refined control strategy underlies the response of fruit flies to in-flight perturbations. To reveal this strategy, we construct a physics-based model of the observed behavioral response."

Body motions are detected by the halteres: "small vibrating organs [. . .] that act as gyroscopic sensors. Anatomical, mechanical, and behavioral evidence indicates that the halteres serve as detectors of body angular velocity that quickly trigger muscle action." With this model, the halteres have a nonlinear response consistent with vibratory gyroscopes, so sensor saturation explains "why fruit flies are unable to accurately recover from strong perturbations". The control system design principles are as follows:

"These findings suggest that these insects drive their corrective response using an autostabilizing feedback loop in which the sensed angular velocity serves as the input to the flight controller. [. . .] [T]he velocity is sensed by the halteres, processed by a neural controller, and transmitted by the flight motor into specific wing motions that generate aerodynamic torque."

Halteres are remarkable organs and unique to the Diptera. The research raises questions about other autostabilization techniques found in the natural world and how such systems can be incorporated into flying robots.

"Flight control principles uncovered in this model organism may also apply more broadly, and this work provides a template for future studies aimed at determining if other animals employ flight autostabilization. The control strategies across different animals are likely to share common features, because the physics of body rotation is similar across many animals during flapping-wing flight. Additionally, animals that lack halteres may use functionally equivalent mechanosensory structures such as antennae. Finally, the control architecture of the fruit fly offers a blueprint for stabilization of highly maneuverable flapping-wing flying machines."

These design principles were incorporated by intelligent agents into aeroplanes very early in their history (further information is here). It is now apparent that flying insects got there first! In evolutionary terms, we have here a good example of convergence. Since these control systems represent complex specified information (with the greater complexity found in the insect control system), intelligent agency should be invoked in both cases.

"For fixed-wing machines, the need to overcome instabilities spurred the invention of autostabilizing systems by 1912, only 9 years after the Wright brothers first manually controlled airplane flight. The development of such automatic steering systems also led to the first formal description of proportional-integral-derivative control schemes and advanced gyroscopic sensor technology. The fruit fly's autostabilization response is well-modeled by a simple PD scheme that receives input from gyroscopic halteres, and, like airplanes, uses fine adjustment of wing orientation to generate corrective torques. Roughly 350 million years after insects took flight, man converged to this solution for the problem of flight control and joined animals in the skies."

Discovering the flight autostabilizer of fruit flies by inducing aerial stumbles
Leif Ristroph, Attila J. Bergou, Gunnar Ristroph, Katherine Coumes, Gordon J. Berman, John Guckenheimer, Z. Jane Wang and Itai Cohen.
Proceedings of the National Academy of Sciences, 2010, 107:4820-4824 | doi:10.1073/pnas.1000615107

Abstract: Just as the Wright brothers implemented controls to achieve stable airplane flight, flying insects have evolved behavioral strategies that ensure recovery from flight disturbances. Pioneering studies performed on tethered and dissected insects demonstrate that the sensory, neurological, and musculoskeletal systems play important roles in flight control. Such studies, however, cannot produce an integrative model of insect flight stability because they do not incorporate the interaction of these systems with free-flight aerodynamics. We directly investigate control and stability through the application of torque impulses to freely flying fruit flies (Drosophila melanogaster) and measurement of their behavioral response. High-speed video and a new motion tracking method capture the aerial "stumble", and we discover that flies respond to gentle disturbances by accurately returning to their original orientation. These insects take advantage of a stabilizing aerodynamic influence and active torque generation to recover their heading to within 2 deg in less than 60 ms. To explain this recovery behavior, we form a feedback control model that includes the fly's ability to sense body rotations, process this information, and actuate the wing motions that generate corrective aerodynamic torque. Thus, like early man-made aircraft and modern fighter jets, the fruit fly employs an automatic stabilization scheme that reacts to short time-scale disturbances.

See also:

Tyler, D. Biorobotics casts light on the way insects fly, ARN Literature blog (22 February 2007)