Introduction

When looking into nature, medium and large sized predatory birds that need to be agile have a distinct functional tail (as opposed to an aesthetic large tail) than most other birds. Examples are eagles, kites, falcons, ospreys and hawks. It’s also interesting that pigeons can outsmart a falcon most times, with their wide field of vision and cunning escape maneuvers for which their tail plays an important role [1]. The most agile northern goshawk that inhabits dense forests in the northern hemisphere manoeuvres through impossibly tight spaces, without hitting anything, in the pursuit of prey. Despite having a wingspan of over a metre, these birds can chase prey at up to 60km/h while manoeuvring zigzag through dense forest. They have relatively flexible wings and larger tails that make them extremely agile. For dodging obstacles they shrink and fold their wings, and uses their tail as a third wing to sustain lift [2]. No human made aircraft can match the agility of the northern goshawk. As a rule of thumb, agility is synonymous to a distinct functional tail among larger bird species.

Adrian Thomas [3] mentioned that the birds could be using the asymmetric orientation of tail for control during turning manoeuvres. The moment arm of the tail about its apex is more than twice the moment arm of a conventional wing about its leading edge, and hence any force would generate a significant moment about its tail. Birds use their tail efficiently and effectively during manoeuvres and hovering, where they do this by change in spread, in angle of attack and in roll relative to the body. Furthermore, they use the tail to produce enough lift during take off and landing, by spreading and drooping the tip feathers. This drooping of tip feathers reduces the induced drag by half and may be used for control as well. This lift generated by the tail is proportional to the square of its maximum span and beyond the point of maximum width drag supersedes lift.

Compared to the aerodynamic function of bird wings, the aerodynamic function of bird tail has been subject of speculation and was never fully agreed upon [3]. The question on why birds excel flight without a vertical tail was always under debate and handful of theories was proposed to substantiate this [3-5]. Sinha and Ananthkrishnan [5] debunked the rationale that vertical tailless flight is achievable only in the realm of scale where birds and miniature aircrafts encompass.

In 1921, Ludwig Prandtl published a paper [6] showing the elliptical spanload as the most efficient wing choice that soon became the standard in aviation. But in 1933 Prandtl published a lesser known second paper [7] proposing a new bell-shaped spanload that is far more efficient and has a longer span than with the elliptical spanload. During roll, the adverse yaw (as a resultant of induced drag) created by a wing based on an elliptical spanload necessitates a vertical tail with rudder. On the other hand, a bell spanload creates a proverse yaw where an auxiliary yaw device like the vertical tail is not essential. This allows for improved aircraft designs, particularly all flying-wing aircrafts and blended-wing body aircrafts [8, 9].

Robert Hoey [10] undertook an effort to determine whether birds are statically stable in soaring flight, and to identify the control method they use while initiating turns. He used a full-scale radio-controlled (RC) glider to model ravens in their soaring flight. The glider tested two tail configurations: (1) a simple elevator at the back of the tail for just pitch control and (2) a more complex “rolling tail” mechanism that allowed the entire tail area aft of the wing to pivot up and down for pitch control and also rotate around a longitudinal axis for a combined pitch and yaw control.

Zheng et. al. [11] in a prepublication paper discusses the application of deep reinforcement learning on a bionic morphing tail for drones. They devised a mechanism that can spread the tail with an area increase of 49.6%, in addition to deflection (up and down at the trailing edge) and rotation (along longitudinal symmetry of the aircraft) of the tail assembly. The variable area mechanism uses the deformation principle of parallelogram. The mechanism is yet to be tested in actual flight conditions on a drone, as inferred from the paper.

Very recently scientists at the Laboratory of Intelligent Systems at EPFL (École Polytechnique Fédérale de Lausanne, Switzerland), led by Prof. Dario Floreano, developed a next-generation drone inspired by the northern goshawk [12 & 13]. They modelled the drone mimicking the bird’s wings and tail, and its flight behaviour. The drone, using foldable wings and a morphing tail, displayed considerable agility and maneuverability.  However, they still kept a vertical stabiliser in the middle of the tail assembly for stability that undermine the actual mimicking from the nature’s finest creations.