Robot pigeon promises new flying machines

As early aviators discovered, sometimes to their terminal cost, the way in which birds fly is complicated and complex. Fixed wings, the Wright brothers and their ilk worked out, were a more practical, if much less nimble, alternative.

The dream of making machines that can soar and yaw and swoop on flapping wings never quite faded away, and today it is roboticists rather than aircraft engineers who are striving to make it come true.

A pair of papers published this week in two journals reveal that the idea is coming closer to realisation, if only, for the moment, as a proof-of-concept demonstration.

200117 bird robot prev
A composite of the PigeonBot wings. Credit: Lentink Lab / Stanford University

Flying robots with flapping, tilting and independently moving wings may not be that far over the horizon. And the real-life bird species that is providing both the inspiration and the anatomical data to make them possible is not the mighty gliding condor, nor the indefatigable migratory Arctic tern, nor even the tiny hovering hummingbird. It is the common or garden pigeon.

In a paper published in the journal Science Robotics, researchers led by Eric Chang from Stanford University in the US detail the development of a robot pigeon, or what they term a “biohybrid morphing wing with real feathers”, or, for the sake of brevity, PigeonBot.

To construct their proto-robo-wing, the scientists first investigated in fine detail the mechanics and physics of real pigeon wings. They determined that the assemblage, comprising 20 primary and the same number of secondary feathers, is controlled by the transfer of energy through the skeleton. 

Effectively, they realised, pigeons control their flight by using their fingers and wrists.

The next step was to construct an artificial wing skeleton – a couple of them, actually – to which they attached real pigeon wing feathers. The robot fingers and wrists on each wing were controlled using four servo-actuated joints, and the feathers connected to each other using elastic.

The early results, the researchers report, have been encouraging. The wings are able to move through a 42-degree arc, can change orientation rapidly, and can be easily repaired – through a kind of “preening” – when they smack into the wall.

More promising still, Chang and his colleagues found that by sending different sets of instructions to the left and right wings, PigeonBot was able to change direction in mid-air.

In a second paper, published in the journal Science, a group led by Chang’s Stanford colleague Laura Matloff took a closer look at the role feathers play in allowing birds to control their flight direction and speed.

Key to this is the fact that the animals are able to change the effective shape, size and surface area of their wings while they are in the air.

The researchers found that the feat was achieved thanks to microstructures on each feather, which they dubbed “directional Velcro”. The structures – technical name, lobate cilla – function as hook-and-eye mechanisms, locking together as the wing is expanded, and then slipping loose again as the wing size contacts and the feathers slide smoothly beneath each other.

The locking mechanism is essential for flight because it strengthens the extended wing and makes it resistant to turbulence. The researchers note that equivalent structures have been found on almost all species of flying birds – except those, such as owls, which have evolved to fly silently.

Matloff and colleagues studied the manner in which the microstructures worked in the feathers attached to PigeonBot and concluded that they “could inspire innovative directional fasteners and morphing aircraft”.

The PigeonBot making turns using asymmetric wing morphing.
CreDIT: Lentink Lab / Stanford University

Related reading: Bat Bot begins: a new agile flying robot for disaster zones

Please login to favourite this article.