Flapping-wing micro air vehicles

Transcription

1 Flapping-wing micro air vehicles M.F. Platzer & K.D. Jones AeroHydro Research & Technology Associates, Pebble Beach, CA, USA. Abstract This paper presents an overview of the two types of flapping-wing micro air vehicles that have been developed in recent years. Biomimetic designs imitate nature while biomorphic designs are merely inspired by nature, but incorporate other design features not found in nature. A recent successful biomimetic flapping-wing micro air vehicle the Microbat, developed by the Aero- Vironment Company is described. Also, the major design features of a biomorphic micro air vehicle, developed by the authors, are presented in some detail. This design incorporates a fixed wing and two flapping wings arranged very close behind the trailing edge of the fixed wing. 1 Introduction Flapping-wing model aircraft date back at least to 1874, when Alphonse Penaud built a rubberband-powered ornithopter. The great aeronautical pioneer Otto Lilienthal [1] was fascinated by the flight of birds and tried to make them the basis for his flight experiments in the 1890s. The course of aeronautical history has shown that fixed-wing and rotary-wing aircraft have distinct advantages over flapping-wing aircraft. Yet, the challenge of developing flapping-wing air vehicles has continued to attract model airplane enthusiasts and aeronautical engineers. For example, the X-wing canard flapper, built in 1985 by Frank Kieser [2], set a new record for indoor free-flight endurance and variations of this design still hold the record. This highly innovative design is shown in Fig. 1. In 1992 DeLaurier & Harris from the University of Toronto [3] successfully flew an engine-powered remotely piloted ornithopter for almost 3 min and DeLaurier still hopes to fly a manned ornithopter [4]. The initiative launched by the Defense Advanced Research Projects Agency (DARPA) in 1996 to develop micro air vehicles has given new impetus to flapping-wing research because flapping wings may be superior to rotary-wing or propeller-driven air vehicles at the small scale required by DARPA. For example, propellers generate torque and create helical slipstreams, both of which degrade vehicle performance and handling. Moreover, the efficiencies of small-scale propellers and rotors are adversely affected by the flow separation effects encountered at low doi: / /5b

2 394 Flow Phenomena in Nature Figure 1: Frank Kieser s X-wing flapper [2]. Reynolds numbers. Flapping wings have larger actuator areas and therefore lower actuator loadings, thus enabling better efficiencies. In principle, one distinguishes two types of flapping-wing vehicles, i.e. biomimetic and biomorphic designs. Biomimetic designs imitate nature while biomorphic designs are merely inspired by nature, but incorporate other design features not found in nature. 2 Biomimetic designs Until Frank Kieser s design in 1985, virtually all flapping-wing fliers were biomimetic. The AeroVironment Company in Simi Valley, CA, took this one step further by exchanging the rubber band motor for an electric motor, producing the elegant Microbat, shown in Fig. 2. The latest version of the Microbat uses a two-cell lithium polymer battery, a three-channel radio, and has a 23 cm span and total weight of 14 g; it has made 25 min flights. While many aeromodeling enthusiasts had previously built successful gas and electric-powered ornithopters, the Microbat seems to be the smallest radio controlled biomimetic flapper. 3 Biomorphic designs As already mentioned, Frank Kieser deserves the credit for a very original design. We took a different approach in our design, shown in Fig. 3. It consists of a fixed wing and two flapping wings arranged very close behind the trailing edge of the fixed wing. Clearly, this is a configuration not found in nature, although insect biplanes existed many millions of years ago, albeit without a fixed wing. As noted by Wootton & Kukalova-Peck [6], the Homoiopteridae were an ancient group of large, sometimes gigantic insects which had forewings and hind wings that overlapped extensively. The same appears to have been the case for some members of the family Lycocercidae. No such overlapping is found any longer in modern insects, leading to the interesting question for the reason for the disappearance of biplane insects.

3 Flapping-Wing Micro Air Vehicles 395 Figure 2: AeroVironment s Microbat [5]. Figure 3: Authors model with a fixed wing and two flapping wings. Our own reason for choosing the biplane configuration with wings that flap in counterphase was twofold. First, the common center of gravity of wings that are flapping in counterphase remains unaffected by the flapping motion of the individual wings in contrast to the use of single wings. Hence the vehicle is dynamically balanced and pitch oscillations of the whole vehicle are eliminated. This provides an inherently more stable platform for use as a surveillance vehicle. Birds have evolved to compensate for these oscillations and their neck is highly articulated such that their head is inertially stable for improved vision. The second equally important reason for choosing the biplane configuration stems from the aerodynamic benefits that accrue from its use. The lift enhancement that is caused by flying a fixed wing close to the ground is well known.

4 396 Flow Phenomena in Nature Similarly, birds flying close to the water surface achieve an increase in thrust and propulsive efficiency by flapping their wings in the ground effect. Using the panel code described in reference [7] these benefits can be quantified and are shown in Fig. 4. The thrust coefficient and the propulsive efficiency are plotted in the upper and lower figures, respectively, as a function of reduced frequency (defined as the product of the circular frequency of flapping and the wing chord divided by the flight speed). It is readily seen that flight in the ground effect, or equivalently flight with two wings flapping in counterphase, is quite beneficial. Furthermore, birds have no choice but to flap their wings such that the flap amplitude varies along the span from zero amplitude close to the body to maximum amplitude at the wing tips. Hence the inner part of the wing contributes little to thrust generation, but the bird has no alternative. A wing that flaps with a constant amplitude along the span would seem to be preferable. Therefore, we chose the arrangement shown in Fig. 3 which makes it possible to drive the wings with constant amplitude along the span. Finally, we decided to make use of the fact that flapping wings act as two-dimensional propellers. Arranging conventional propellers along the trailing edge of a wing energizes the flow over the upper wing surface because of the propellers suction effect. As a consequence, wing stall is delayed to a higher angle of attack than possible with a more conventional arrangement. In contrast to conventional propellers, flapping wings are ideally suited for such an arrangement. An additional argument for its use was the greater tendency toward flow separation in micro air vehicles because of the much lower flight Reynolds numbers encountered in such vehicles (typically 30,000 for the main wing in the model shown in Figure 3). We explored the effectiveness of flapping wings in suppressing flow separation in a series of water tunnel tests. The cusped trailing edge of an airfoil together with a small airfoil mounted close to the trailing edge is shown in Fig. 5. This small airfoil could be oscillated in plunge. The flow over the cusped trailing edge is fully separated due to the large adverse pressure gradient in this region. However, the flow could be made to reattach by the plunge oscillation of the small airfoil, as shown in Fig. 5. Therefore, we decided to mount the two flapping wings very close to the trailing edge of the fixed wing. Figure 4: The predicted benefits of the ground effect [8].

5 Flapping-Wing Micro Air Vehicles 397 It is seen that this decision process led us to a vehicle configuration that is truly biomorphic. At first appearance it would seem that the design separates the lift and thrust generators, as is commonly done in conventional aircraft. However, all three wings form a system that together efficiently produce lift and thrust, but their efficiency depends on mutual interference. The biplane flapping wings need each other in order to get the benefit of flight in the ground effect, and the main wing needs the downstream flapping wings in order to maintain the attached flow for efficient lift. There are perhaps other bio-inspired facets of the design. For example, many birds and insects do not actively control the angle of attack or twist of the wings, but rather they are deflected merely by aerodynamic loading an aeroelastic deflection. Similarly, the flapping wings on our model are connected to the flapping mechanism with tiny elastic members which allow for passive aeroelastic pitching. A wind tunnel variant of the model in Fig. 3 was built and is shown in Fig. 6. Using flow visualization with a smoke wire, the effectiveness of the flapping wings in suppressing flow separation is seen in Fig. 7, where the wings are not flapping in the image on the left, with massive flow separation from the leading edge, and with the wings flapping in the image on the right, the flow is seen to be almost completely attached. Impressively, the video frame shown on the right is just four frames after flapping started, with a flapping frequency of about 30 Hz, which corresponds to four flapping strokes or about 1/8th of a second to transition from fully separated to fully attached flow. Figure 5: Flow reattachment due to airfoil oscillation [8]. Figure 6: The wind tunnel test model [8].

6 398 Flow Phenomena in Nature Figure 7: Flow separation control by means of flapping wings [8]. 4 Design and development of the fixed/flapping-wing micro air vehicle For the implementation of the above-described design concepts into a viable flying vehicle the energy source and the motor are critical issues. Fortunately, the requirements of the cell phone industry for batteries of high energy density and for tiny electric motors (so-called pager motors the motors which vibrate cell phones and pagers) pushed the state of the art to the point where it became feasible to start assembling a flight vehicle. In late 2002, the available lithium polymer cells weighed 4.2 g, the motor/gear assembly about 2.5 g, the DC DC step-up circuit needed to increase the battery voltage to 5 V about 0.7 g and the three-channel receiver for the radio gear about 2 g. The weight of the structure itself was estimated to be about 5 g, yielding a total weight of about 14 g. Using the best aerodynamic data for wings of this scale, a main wing with a span of 30 cm and a chord of 14.5 cm was built. The first flight took place in December 2002, lasting about 3 min, when the unguided model landed high in a tree. The flight speed was only about 2 m/s, but the ability of the micro air vehicle to fly at very high angles of attack without stalling fully confirmed the effectiveness of the above-described flow separation control method. With the power off, the micro air vehicle stalled quite easily in response to gusts, but turning the flapping wings on restored control right away. A second micro air vehicle was built a few months later, which had a slightly smaller span of 27 cm and a weight of 13.4 g. The third micro air vehicle was still slightly smaller with a 25 cm span and a weight of 12.4 g. The smallest micro air vehicle, built for the 8th International MAV Competition, had a span of 23 cm and a weight of 10.5 g. The flight speeds of all the models varied between 2 and 5 m/s, and battery capacities allowed for flights up to 20 min duration. The very low flight speed made the model ideal for flight in confined areas. We routinely flew them in lecture halls as part of our presentation. 5 Summary and outlook Stimulated by DARPA s initiative in 1996 significant progress has been achieved in the development of micro air vehicles. In this paper we have described two flapping-wing micro air vehicles which have reached flight status, i.e. the biomimetic Microbat, designed and developed by the AeroVironment Company, and our biomorphic fixed/flapping-wing micro air vehicle. Both vehicles benefited from the advances made in the fixed and flapping-wing aerodynamics for micro air vehicle applications, as reviewed in reference [9]. However, both developments were still largely based on cut and try methods. Hence, much more experimental and computational work in the

7 Flapping-Wing Micro Air Vehicles 399 field of low Reynolds number aerodynamics is required in order to optimize future designs. The amazing recent advances in battery and motor technology, driven by the cell phone industry, have made it possible to achieve success. These advances, together with the further miniaturization of the avionics components, are certain to continue. Hence the design and development of new micro air vehicles is likely to pose many further exciting challenges for the aeronautical engineer. References [1] Lilienthal, O., Der Vogelflug als Grundlage der Fliegerkunst, R. Gaertners Verlagsbuchhandlung: Berlin, [2] Ornithopter Media Collection, published on the web at indoor.shtml [3] DeLaurier, J.D. & Harris, J.M., A study of mechanical flapping wing flight. Aeronautical Journal, 97(968), pp , October [4] Larijani, R.F. & DeLaurier, J.D., A nonlinear aeroelastic model for the study of flapping wing flight (Chapter 18). Progress in Astronautics and Aeronautics, Vol. 195, American Institute of Aeronautics and Astronautics, [5] Keennon, M.T. & Grasmeyer, J.M., Development of the black widow and microbat MAVs and a vision of the future of MAV design, AIAA/ICAS International Air and Space Symposium and Exposition The Next 100 Years, AIAA , July [6] Wootton, R.J. & Kukalova-Peck, J., Flight adaptations in Palaeozoic Palaeoptera. Biological Review, 75, pp , [7] Jones, K.D., Dohring, C.M. & Platzer, M.F., Experimental and computational investigation of the Knoller-Betz effect. AIAA Journal, 36(7), pp , [8] Jones, K.D., Bradshaw, C.J., Papadopoulos, J. & Platzer, M.F., Improved performance and control of flapping-wing propelled micro air vehicles, AIAA , January [9] Mueller, T.J., Fixed and Flapping Wing Aerodynamics for Micro Air Vehicle Applications, Progress in Astronautics and Aeronautics, Vol. 195, American Institute of Aeronautics and Astronautics, 2001.