I researched winged-flight in robots, focusing on hovering, and further focusing on hovering. A prime example of hovering in the animal world is that of dragonflies.

I will ask what a robot requires to hover in the air such as a dragonfly does.

C1.0) Source

The paper [1] delves into this very issue. Dragonflies are not just great at hovering; [1] venerates the dragonfly as “one of the most agile and maneuverable flying insect species.” Unlocking the secret of the dragonfly wings will naturally lead to additional scientific advances regarding robotic flight.

The authors of [1] build wings that seek to imitate that of a dragonflies and move them and attempt to subject them to forces that would be felt in-flight and while hovering, and measure the results. Particularly, the authors focus on replicating the dragonflies’ high-stroke plane and drag based lift generation.

C1.1) Venue

The venue of the paper is of quality. The paper was presented at the prestigious 2009 IEEE International Conference on Robotics and Automation. Taking place in Japan, the conference was put on by the Institute for Electric and Electronic Engineers (IEEE), an organization founded in 1963 and today having more than 400,000 members. IEEE’s mission statement is to “foster technological innovation and excellence for the benefit of humanity” (www.ieee.org).

C1.2) Authors

Zheng Hu is a mechanical engineering masters student at the University of Delaware. Raymond McCauley is another mechanical engineering student who won an award from NASA in 2007 for research done.

Xinyan Deng received a PhD in Mechanical Engineering from the University of California at Berkeley. He received an NSF career award in 2006, and his papers have appeared in Science and his research covered by major newspapers. Scopus gives him an h-index of 7.


I began my search by going to Scopus and searching for “flight AND wing AND robot”. I performed the same search at the ISI Web of Science. After I decided to focus on hovering, I searched for “wing AND hovering” from 2005 to 2011 in the ISI Web of Knowledge. I found about 12 sources from these search results (and from following a couple of references) that I felt interested in using, though some of them were from too long ago.

I wrote of my sources and search techniques to the library consultant. He identified a few great sources for me to look at, one of which I have used as one of my main sources. He also helpfully explained how he found his sources, in a search on Scopus and Compendex with the search “hover* and wing* and robot*”. He provided an important reminder to use wildcards. He also commented on what had been my choice for main article, pointing out how that particular journal was not as high quality as I had originally supposed.

After that I had almost enough sources, although I did do a search on PubMed for “dragonfly hovering” which turned up [8] and [9].

C1.4) High Quality Bioinspired Robotics Contribution

Robots have not been built that utilize dragonfly wings to hover. There have been tests done on some wings and theoretical models to move slightly in this direction.

[1] takes it further by physically building wing that is similar to a dragonfly wing, which the authors subject to forces and experimentation.

The authors say that, “The results will contribute to the design of micro aerial vehicles with two pairs of wings.”

C1.5) General Robotics Literature

In 2005, attempts at creating a hovering robot had not quite reached the attempt stage. One paper, [2], documents the efforts to build a wing system that would move in a way that it was thought wings should move to enable hovering, and would be subjected to the same forces, and documents the resulting measurements and data. The goal was not to have an actual hovering robot, but to simulate hovering (albeit with physical wings).

[2] is published by the Journal of the Royal Society Interface, which has an impact factor of 4.241 for 2009. It is ranked 4th in JCR’s multidisciplinary category. The authors are Cezary Galiński and Rafał Żbikowski, who have h-indexes of 4 and 10, respectively.

As [2] explains, in hovering, downstroke and upstroke of the wings are equal and wingtips trace a shape akin to a figure of eight. The “flapping mechanism” was built on the scale of an insect and with wing-beat frequency of 20 Hz, and motion of each wing a mirror-image of the other. Other specifications were made. All self-assigned design criteria were met, and the machine works. One must wonder, of course, how close this presumed hovering mechanism is to the design which imitates nature to a near enough degree so as to enable successful hovering in reality. (Indeed, the authors in [9] point out after observing a dragonfly that the dragonfly employs an inclined stroke plane. The apparatus in [2] has a stroke plane parallel to the body. As an additional comment, [9] was published a year before [2].)

Before a creature or robot hovers, it must lift off from the ground. Liftoff is dealt with by [3], a paper found on Scopus.

It is a paper from the 2007 IEEE International Conference on Intelligent Robots and Systems. The author, R.J. Wood, has an h-index of 4, which seems to leave something to be desired until one takes into account that he has only been publishing since 2005. In that time he has been published 13 times and has been cited 49 times. Wood graduated from Harvard School of Engineering and Applied Science.

In [3], Wood builds a winged robot that generates sufficient force to lift itself off the ground. It does so, however, with the help of guide wires that constrain movement such only vertical change is permitted. Even taking into account this caveat, this paper represents the first time that “researchers working to artificially re-create hovering insect flight have demonstrated…an integrated micromechanical device that is able to lift its own weight.”

C1.6) Biology Literature

Previously, [1] noted the complexities of a dragonfly’s wing. A newer paper, [4], examines the structure of a Dragonfly wing in greater detail.

The journal in which [4] is published, Experimental Mechanics, has a Scopus Source Normalized Impact Per Paper of 1.720. They also have an impact factor of 1.542, and rank 5th in the ISI Category of Materials Science, Characterization and Testing. It is also an official journal for the Society of Experimental Mechanics, founded in 1943. Although the article has not been cited yet, it was only written in November of 2010. The authors of [4] are S.R. Jongerius and D. Lentink. Lentink has an h-index of 6.

The authors of [4] collect a dragonfly wing from a dragonfly and subject it to tests and measurements. Subsequently, they create a “three-dimensional structure and load model of a dragonfly wing.” This will be useful for simulations. Thickness of the wing veins and membrane decreases from the root to the tip of the wing, which aids in supporting inertial and aerodynamic loads. Inertial loads are the stronger of the two, and deform the wing most during stroke reversal. Dragonflies flap their forewings at frequencies about 5 times lower than the first natural vibration mode of the forewing, a method of motion that differs from the typical motion of “advanced” micro-air vehicles (MAV) built to date, which run at much higher frequencies. (The authors consider advanced micro air vehicles to be those found in [5], [6], and [7].) On a possibly related note, the dragonfly’s wing is stiffer than that used in MAVs as well. The authors posit that in the current deformable, more energy is lost, although their slackness makes them easier to actuate. They suggest “stiffer and more precisely designed” wings for superior MAVs modeled after the dragonfly.

Wang and Russell examine a live dragonfly in [8]. It is published in Physical Review Letters, which has an impact factor of 7.328. David Russell has an h-index of 40 and Zhi Jane Wang has an h-index of 19.

The authors of [8] tether a live dragonfly and capture its wing movements on video. The researchers are paying particular attention to how the dragonfly modulates the phase lag between forewings and hindwings. (Incidentally, in [2] there is no such wing independence.) During takeoff the wings beat in phase, and during hovering the wings beat out of phase. The counterstroking behavior during hovering reduces force fluctuations and body oscillation, and more importantly it minimizes the power consumed. It allows the dragonfly to expend less energy to hover. This is certainly a trait that would be useful for robots to adopt!

Indeed, the dragonfly is the model of the future. Understanding its secrets and incorporating them into MAVs will allow robotics to soar to new heights.


Numbers are not in order found, but generally in order used in the assignment.

1. Zheng Hu, Raymond McCauley, Steve Schaeffer, and Xinyan Deng // Aerodynamics of Dragonfly Flight and Robotic Design.// 2009 IEEE International Conference on Robotics and Automation, May 12-17, 2009
2. Cezary Galiński and Rafał Żbikowski // Insect-like flapping wing mechanism based on a double spherical Scotch yoke // Journal of the Royal Society Interface, 2005, (2), pp. 223-235
3. Wood, R.J // Liftoff of a 60mg flapping-wing MAV // IEEE International Conference on Intelligent Robots and Systems, Article number 4399502, Pages 1889-1894, 2008.
4. S.R. Jongerius and D. Lentink // Structural Analysis of a Dragonfly Wing // Experimental Mechanics, 2010; 50, (9), pp. 1323–1334.
5. Pornsin-Sirirak TN, Tai YC, Ho CH and Keennon M (2001)
Microbat-a palm-sized electrically powered omithopter. NASA/ JPL Workshop on Biomorphic Robotics, Pasadena, USA.
6. Kawamura Y, Souda S, Nishimoto S and Ellington CP //
Clapping-wing micro air vehicle of insect size // ed. Kato N and Kamimura S; Bio-mechanisms of swimming and flying. Springer Verlag, 2008 pp 319–330
7. Floreano D, Zufferey JC, Srinivasan MV //The scalable design of flapping micro air vehicles inspired by insect flight //
8. Z. Jane Wang and David Russell // Effect of Forewing and Hindwing Interactions on Aerodynamic Forces and Power in Hovering Dragonfly Flight // Physical Review Letters, 2007, Oct 5; 99, (14) article number 148101
9. Wang ZJ // The role of Drag in Insect Hovering // Journal of Experimental Biology, 2004, 207, (23) pp. 4147-4155
10. DiLeo C, Deng XY // Design of and Experiments on a Dragonfly-Inspired Robot // Advanced Robotics, 2009; 23, (7-8), pp. 1003-1021
11. Wang, Z.J. // Dissecting Insect Flight // Annual Review of Fluid Mechanics, 2005, 37, pp. 183 – 210
12. Fearing, R.S., Chiang, K.H., Dickinson, M.H., Pick, D.L., Sitti, M., Yan, J // Wing transmission for a micromechanical flying insect // IEEE International Conference on Robotics and Animation, 2000. Volume 2, 1509-1516
13. Tanaka, H., Matsumoto, K., Shimoyana, I. // Design and Performance of Micromolded plastic butterfly wings on butterfly ornithopter.
14. Steltz, E., Avadhanula, S., Fearing, R.S. // High Life Force with 275 Hz wing beat in MFI // IEEE International Conference on Intelligent Robots and Systems, 2007, Article number 4399068, Pages 3987-3992
15. Tobalske BW // Hovering and Intermittent flight in birds // Bioinspiration and Biomechanics, 2010, 5, (4), article number 045004