Welcome to The Neuromorphic Engineer
Applications » Locomotion

Wing-wing interactions in dragonfly flight

PDF version | Permalink

Xinyan Deng and Zheng Hu

18 November 2008

A pair of robotic wings have been used to investigate why dragonflies use different phase angles for different flight modes.

The dragonfly is one of the Earth's most maneuverable insects and oldest species. Its flight performance far exceeds other flying insects: it can hover, cruise up to 54km/h, turn 180° in three wing beats, fly sideways, glide, and even fly backwards. Dragonflies intercept prey in the air with amazing speed and accuracy. To achieve this, most change their wing motion kinematics for different flight modes such as hovering and turning. The most noticeable of these changes is the phase difference between fore- and hind wings, defined as the phase angle by which hindwing leads the fore. When hovering, dragonflies employ a 180° phase difference (out of phase),1–3 while 54–100°is used for forward flight.4,5 When accelerating or performing aggressive maneuvers, there is no phase difference between the two wings (0°, in phase).1, 3,6 Interestingly, among the various flight modes, 270° phase difference is rarely observed in dragonflies.

A dragonfly in motion (viewed from right).

Besides having two pairs of wings for added lift force and maneuvering control, dragonflies employ an inclined stroke plane in which the wing motion is mostly confined (see Figure 2). While most flying insects use a horizontal stroke plane, dragonflies' are approximately 60° from the horizontal.1–3 Their wings act as if ‘paddling’ in the air in the sense that the chord is almost horizontal during downstroke to generate the maximum upward force and is close to being vertical during upstroke to reduce the downward force. Therefore, their aerodynamic mechanism is ‘drag base lift generation’ as shown in computational fluid dynamics (CFD) studies.7 CFD results also showed that the effect of wing-wing interaction is actually detrimental to lift force generation in dragonflies.8, 9

Sketches of the wing chord snapshots in hover mode (viewed from left). Circles indicate the leading edges.

To simulate dragonfly motion during hovering and forward flight (see Figure 3) we constructed a pair of dynamically-scaled robotic wings. Briefly, we use a pair of bevel-geared robotic wrists to generate rotational motion in co-axial roll-pitch-yaw degrees of freedom. The stroke planes of the flappers can be arbitrarily changed between 0–90° and the whole apparatus is mounted on a linear stage driven by a stepper motor to study forward motion. Meanwhile, a six-channel force/torque sensor was mounted at the wing base to measure instantaneous aerodynamic forces.

For those interested in the technicalities, the robotic wings are powered by 16mm, 0.3Nm torque DC brush motors equipped with magnetic encoders to provide kinematic feedback. These are driven along kinematic patterns provided by a custom MATLAB Simulink program with WinCon software that provides commands to the real-time-control and data-acquisition board communicating with the hardware. We use proportional-integral-derivative (PID) controllers to run the motors with precision of 0.1°. Motion commands from the computer are amplified by analog amplifier units that directly control the input current received by the motor.

In the experiments, we employed the real dragonfly (Aeshna juncea) kinematics from biological data2 and systematically varied kinematic parameters such as the phase differences between the forewing and hindwings, the forward speed and the distance between the wing bases. The effect of phase difference in hover and forward flight is shown in Figure 4.

Experimental setup schematics (left) and robotic wrists (right).

Total lift force coefficients results from hovering (A) and forward flight (B). The straight lines in each plot stand for the total lift force from both wings without wing-wing interaction.

It is interesting to see an overall detrimental effect to lift force generation due to wing-wing interactions. Furthermore, when the phase difference is around 0° (in phase), the total lift force tends to be higher than other cases, and it gets lower around 180° in hover mode and 270° for forward flight. Our results proved that in-phase flight generates larger aerodynamic forces than out-of-phase flight or single wings added together. This explains why dragonflies use in-phase flight for aggressive maneuvers such as turning or accelerating. While out-of-phase motion is detrimental to force generation, dragonflies use it for hovering stability and vibration suppression. This makes sense, as we have observed that out-of-phase and in-phase stroking produce regular and irregular flight respectively.3 The results also help explain why dragonflies never favor the 270° phase difference.

The study not only expands our understanding of biology, but also gives us an indication of how to build a four-wing micro air vehicle. Specifically, engineers will need to consider how to coordinate the motions of the two wing pairs when the requirements are either best aerodynamic effect or best dynamical stability. In the first case energy is conserved to some extent, which is good for long distance flight. The second case vibration—which is detrimental to performance in tasks such as inspection and detection—is reduced. The latter also keeps the aircraft safe when flying in turbulence.

Eventually we would like to build a dragonfly-like aircraft. But advanced manufacturing techniques and equipment will be critical to make such a tiny robot possible. In addition, we will have to find a better energy source than currently exists: a light-weight but high-power battery is required. As progress is made in these areas, so we would hope to make progress with our dragonfly.


Xinyan Deng
Mechanical Engineering, University of Delaware

Zheng Hu
Mechanical Engineering, University of Delaware

  1. D. E. Alexander, Unusual phase relationships between the forewings and hindwings in flying dragonflies, J. Exp. Biol. 109, pp. 379-383, 1984.

  2. R. A. Norberg, Hovering flight of the dragonfly: Aeschna juncea L., kinematics and aerodynamics, Swimming and Flying in Nature 2, pp. 763-780, 1975.

  3. G. Rüppell, Kinematic analysis of symmetrical flight manoeuvres of odonata, J Exp Biol 144, pp. 13-42, 1989.

  4. A. Azuma and T. Watanabe, Flight performance of a dragonfly, J. Exp. Biol. 137, pp. 221-252, 1988.

  5. H. Wang, L. Zeng, H. Liu and C. Yin, Measuring wing kinematics, flight trajectory and body attitude during forward flight and turning maneuvers in dragonflies, J. Exp. Biol. 206, pp. 745-57, 2003.

  6. A. L. Thomas, G. K. Taylor, R. B. Srygley, R. L. Nudds and R. J. Bomphrey, Dragonfly flight: free-flight and tethered flow visualizations reveal a diverse array of unsteady lift-generating mechanisms, controlled primarily via angle of attack, J. Exp. Biol. 207, pp. 4299-323, 2004.

  7. Z. J. Wang, The role of drag in insect hovering, J. Exp. Biol. 207, pp. 4147-55, 2004.

  8. H. Huang and M. Sun, Dragonfly forewing-hindwing interaction at various flight speeds and wing phasing, AIAA J. 45, pp. 508-511, 2007.

  9. Z. J. Wang and D. Russell, Effect of forewing and hindwing interactions on aerodynamic forces and power in hovering dragonfly flight, Phys. Rev. Lett. 99, 2007.

DOI:  10.2417/1200811.1269


Tell us what to cover!

If you'd like to write an article or know of someone else who is doing relevant and interesting stuff, let us know. E-mail the editor and suggest the subject for the article and, if you're suggesting someone else's work, tell us their name, affiliation, and e-mail.