ShyyLab :: Research

Micro Air Vehicles

Introduction

Micro air vehicles (MAVs), with a maximal size of 15 cm and flight speed around 10 m/s, have gained growing interest from both military as well as civilian community. Equipped with a video camera or a sensor, these vehicles can perform surveillance and reconnaissance, targeting, and bio-chemical sensing at hazardous location. With the rapid progress made in structural and material technologies, miniaturization of power plants, communication, visualization, and control devices, numerous groups have developed successful MAVs, see Figure 1.

MAVs provide rich research topics, including low Reynolds number aerodynamics, fluid and structure interaction, and unsteady aerodynamics, for both engineering and science societies. Based on the wing type, MAVs can be generally categorized in three groups: fix-wing, rotary wing, and flapping wing. Our group has actively involved in fixed wing and flapping wing study in the past decade. In the following we briefly present the challenges involved based on wing type and the current state-of-art results.

mavs_717

Figure 1. Representative MAVs. (a) flexible fixed wing; (b) rotary wing; (c) hybrid flapping-fixed wing, using fixed wing for thrustl and (d) flapping wing for both lift and thrust.

Fixed Wing - Low Reynolds Number and Low Aspect Ratio Wing

Because of its small size and low flight speed, an MAV operates under low Reynolds number conditions (10⁴-10⁵). It is well known that as the Reynolds number drops from 10⁶ to 10⁵, the aerodynamic performance drops dramatically due to flow separation. Low Reynolds number airfoils demonstrate characteristics different from high Reynolds number ones. First, the low Reynolds number condition favors a thin airfoil with modest camber; if the Reynolds number decrease further, an airfoil with rough surface will be better than a smooth one. Second, the low Reynolds number airfoil may show zigzag behavior in the lift-drag polar, see Figure 2. The zigzag is caused by laminar separation bubble and the laminar-to-turbulent transition. We coupled a RANS equation with the eN method to study the laminar-to-turbulent transition. Our results show good agreement with the experimental measurement in terms of separation position, transition position, reattachment point, and vortex core, see Figure 3.

Figure 2. Lift-drag polar contours of Eppler E374 airfoil, exhibiting substantial Reynolds number effect and the qualitatively different drag bulge as the lower Reynolds number is lowered.

lamturb_560

Figure 3. Streamlines and normalized shear stress contours. Left (AoA = 4 degrees): Top: experiment and bottom: CFD; Right (AoA = 8 degrees): Top experiment and bottom: CFD.

In MAV design practice, the wing typically has a low aspect ratio and has a strong presence of tip vortex. The tip vortex effects are twofold: first, it causes downwash that decreases the effective AoA and increases the drag; second, it creates a low-pressure zone on the top surface and enhances the lift. Figure 4 shows the streamlines and vortices on such a wing at angle of attack of 39 degrees. The low pressure zone is shown in Figure 5. The lift slope of the low aspect ratio wing is lower than that of high aspect ratio wing. However, because of the tip vortex effect, low aspect ratio wing usually does not experience sudden stall.

Figure 4. Streamlines and vortices for rigid wing at 39°.

rigidpressure39_569

Figure 5. Pressure distribution around the rigid wing in the cross sections with streamlines at angle of attack of 39°.

Fixed Flexible Wing - Fluid and Structure Interaction

Flight test, experiment, and numerical tests show that a flexible membrane wing is better than a rigid wing in that it can, first, delay the stall angle, second, give high lift-to-drag ration at higher AoAs, and third, provide a more stable platform in gust environment. It is further noticed that the flexible membrane wing passively adjust its shape according to the flight environment, resulting in a fluid and structure interaction problem, see Figures 6 and 7.

wingstiffness_560

Figure 6. Aerodynamic parameters vs. angle of attack for configurations with varying wing stiffness. (a) Lift coefficient versus angle of attack; (b) Lift to drag ratio versus the angle of attack.

liftdragvstime_530

Figure 7. Experimental lift to drag ratio results for rigid, flexible and hybrid wing at Re=7.5x10⁴ and angle of attack at 7°. The latex membrane wing exhibits about 6% camber at 35.4 fps. The hybrid wing has a curved wire screen camber stop.

Flapping Wing Aerodynamics

Flapping wing aerodynamic phenomena prominently features unsteady motions, characterized by large-scale vortex structures, complex flapping kinematics, and flexible-wing structures.

Under certain combination of plunging frequency and amplitude, a plunging airfoil can generate thrust. The vortex pair shed from the trailing edge tilts towards downstream direction, the so called reverse Karman vortex street (Figure 8).

For an airfoil experiencing both plunging and pitching motion, the leading edge vortex enhances both the lift and thrust (Figure 9).

karmanvortex_866

Figure 8. Streakline plot of flow over plunging airfoil NACA0012 at Re=20000 and reduced frequency of 3.93.

vorticitycontours_536

Figure 9. Vorticity contours at different instants during one flapping cycle: h₀=0.75, St=0.3, k=0.63, ψ=75°, and α₀=15°.

Many biological insects or animals employ normal hovering modes in which the wing moves in a horizontal plane. Along with another popular hovering mode - the water treading mode - experimental and numerical tests investigate the interplay between the hovering kinematics and the fluid physics to gain further insight into the flapping wing aerodynamics. Results show that the main lift generation mechanism for the water treading mode is the delayed-stall, while the wake capturing mechanism is also a contributing factor for the normal hovering mode, see Figures 10 and 11. Hence, depending on the detailed kinematics, the lift generation mechanisms at this Reynolds number exhibit different physical mechanisms.

normalhovering_575

Figure 10. One cycle force history for two hovering modes and quasi-steady value of normal hovering mode at Re_f2=100, h_a/c=1.4, α_a=45°, k=0.357. (a) Lift coeffcient; (b) Drag coefficient.

vorticityhovering_560

Figure 11. Vorticity contours for two hovering modes. Red: counter-clockwise vortices; Blue: clockwise vortices at Re_f2=100, h_a/c=1.4, α_a=45°, k=0.357. The flow snapshots (t1 to t8) correspond to the time instants defined in Figure 10.

References

Researchers:

U of M \| CoE \| Aerospace
Computational Thermo-Fluids @ U of M, Aerospace Engineering