Avian inspired morphing wings
Abstract/Contents
- Abstract
- Bird flight has fascinated humans since ancient times. Early aviators built artificial wings and attempted to fly with limited success. More recently, we have begun to understand the aerodynamics behind bird flight and build our own flying machines. In parallel, biologists have studied the mechanisms and evolution of bird wings and hypothesized how birds first evolved to fly. Both engineers and biologists have noticed that bird wings morph, or change shape, during flight. This morphing enables birds to achieve feats such as efficiency and maneuverability across flight speeds, maneuvering in cluttered environments and flying in turbulent air conditions. Studies have documented the underlying wing skeletal structure and hypothesized that it functions as a "drawing parallels" four bar mechanism. Currently, flying robots are used in applications including package delivery, cinematography, surveillance, and search and rescue. However, further expansion of these applications is limited by the fact that current robots have difficulty flying in real world environments. Most current flying robots are designed with primarily fixed wing shapes and change wing shape only through small movements of control surfaces. Recently, bio-inspired morphing wing designs have incorporated more extreme forms of morphing including large changes in wing sweep, area, twist, and dihedral angle. The goal of this work is to better understand the biomechanics behind how bird wings morph and use that knowledge to improve robotic wing morphing design. These improvements in wing morphing may help robots perform better in real world scenarios. As inspiration for our robotic prototypes, we present three-dimensional kinematic measurements of wing skeletal motion to corroborate the mechanism underlying wing morphing in pigeons (Columba livia). These measurements include three-dimensional tracking of the large wing bones of birds during high-resolution motion capture trials, and tracking of all wing bones including the small wrist bones during seven µCT scans in each of three pigeons. This represents the most thorough dataset on avian wing morphing available. With this dataset, we corroborate potential joint types and derive best fit joints between each pair of adjacent bones. We then analyze 83 potential wing morphing mechanisms to determine how well they represent our data. We look closely at the role of the wing bones and find that the ulnare plays a statistically significant role in changing wing shape and that its effect is seen even in the motion capture data where it is not measured directly. This provides evidence that the previously hypothesized four-bar 'drawing parallels' wing morphing mechanism is lacking. To best fit a four-bar mechanism to the measured data, the radius and ulna bars must cross. Instead, a more complete model incorporating three dimensional motion and wrist bones is needed to understand wing morphing in birds. Inspired by our findings on how birds use their skeletons to morph their wings, we create two wing morphing robots. The first robot passively unfolds its wings in in order to recover from collisions. This robot includes a membrane wing with an underactuated wrist joint allowing free sweep rotation. Therefore, when an obstacle hits the wing it can automatically deflect out of the way. When the obstacle passes and the wing is free to move again, it automatically unfolds due to inertial accelerations induced by flapping. We demonstrate that unfolding is driven by inertial effects through experiments, computational simulations and non-dimensional analysis. Unfolding takes place in approximately the same number of wingbeats across size scales, suggesting it works equally well for both slow and fast flapping wings. The relevant factors driving unfold time are the ratio of the fully extended wingspan to the fully retracted wingspan and the flapping amplitude, with larger flapping amplitudes resulting in unfolding within a single wingbeat. Finally, we show that the passive unfolding documented in this robot also applies within about one wingbeat to birds, bats and insects, which reduces both the neural control and muscle power needed to unfold their wings. The second robot is a bio-hybrid flying robot that actively morphs its wings using a mechanism based on the pigeon data and a feathered morphing surface. It represents both the first feathered bio-hybrid flying robot as well as the first robot to fly with a large (40) number of morphing wing elements. We perform a principle component analysis of the data from the pigeon experiment, demonstrating that a single degree of freedom explains about 84% of the measured motion. We find that the primary contributor to this degree of freedom is wing sweep angle. Therefore, we create a mechanism which includes one actuator for each wing and moves the wing through poses designed to replicate those achieved by pigeons during gliding flight. We used 40 pigeon feathers to form the surface of the wing, forming an underactuated aerodynamic surface constrained by connective springs and feather friction. We present analysis on how these wing shape changes will affect the feathers and the aerodynamic properties of the wing. By prototyping subsequently more complex feathered morphing robots and testing them in flight, we show that using real pigeon feathers is a promising approach to test morphing wing effectiveness in robotico. This represents the first robot capable of flight with such large number of independent wing elements. It is also the first biohybrid aerial robot. Together these measurements and robots represent a thorough exploration of how wing morphing works and demonstrate robotic applications through functioning prototypes. With further development and refinement, these wing morphing mechanisms can be used to improve future flying robots.
Description
| Type of resource | text |
|---|---|
| Form | electronic; electronic resource; remote |
| Extent | 1 online resource. |
| Publication date | 2017 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Associated with | Stowers, Amanda Kay |
|---|---|
| Associated with | Stanford University, Department of Mechanical Engineering. |
| Primary advisor | Lentink, David, 1975- |
| Thesis advisor | Lentink, David, 1975- |
| Thesis advisor | Cutkosky, Mark R |
| Thesis advisor | Delp, Scott |
| Advisor | Cutkosky, Mark R |
| Advisor | Delp, Scott |
Subjects
| Genre | Theses |
|---|
Bibliographic information
| Statement of responsibility | Amanda Kay Stowers. |
|---|---|
| Note | Submitted to the Department of Mechanical Engineering. |
| Thesis | Thesis (Ph.D.)--Stanford University, 2017. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2017 by Amanda Kay Stowers
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...