BionicFlyingFox

Ultra-light flying object with intelligent kinematics

For the BionicFlyingFox, our developers from the Bionic Learning Network took a close look at the flying fox and technically implemented its special flight characteristics. The interaction of the integrated on-board electronics with an external motion tracking system enables the ultra-light flying object to move semi-autonomously in a defined airspace.

The flying fox belongs to the order of bats – the only mammals that can actively fly. A special characteristic is its fine and elastic patagium (flying membrane), which extends from the extended middle hand and finger bones to the ankles. When flying, the animals use their fingers to control the curvature of the patagium and can thus move through the air in an aerodynamic and maneuverable manner. This gives them maximum lift, even during slow flight maneuvers.

Agile kinematics based on a natural role model

With a wingspan of 228 cm and a body length of 87 cm, the artificial flying fox weighs only 580 grams. Like the natural flying fox, its wing kinematics are also divided into arm and metacarpus and covered with an elastic membrane that extends from the wings to the feet. As a result, its wing area is comparatively large and allows low wing loading. As with the biological role model, all pivot points are in one plane, so that the BionicFlyingFox can control and fold its wings individually.

Specially developed patagium

The patagium of the model is wafer-thin, ultra-light and at the same time robust. It consists of two airtight films and an elastane knit which are welded together at approximately 45,000 points. Due to its elasticity, it remains virtually wrinkle-free even when the wings are retracted. The honeycomb structure of the knitted fabric prevents small cracks in the patagium from spreading. This allows the BionicFlyingFox to continue flying even if the fabric is slightly damaged.

Semi-autonomous flying in a defined space

The BionicFlyingFox communicates with a so-called motion tracking system so that it can move semi-autonomously within a defined airspace. The installation permanently detects its position. At the same time, the system plans the trajectories and provides the necessary control commands. The human pilot performs take-off and landing manually. An autopilot takes over in flight.

Movable camera system for exact localization

Two infrared cameras sitting on a pan-tilt unit are an important component of the motion tracking system. This allows them to be turned and tilted so that they can follow the entire flight of the BionicFlyingFox from the ground. The cameras recognize the flying fox by means of four active infrared markers which are attached to the legs and the wing tips.

Machine learning of the ideal trajectory

The images from the cameras are sent to a central host computer. It evaluates the data and coordinates the flight like an external air traffic controller. For this purpose, the BionicFlyingFox has pre-programmed paths on the computer which determine the flight path during its maneuvers. The artificial flying fox uses its on-board electronics and complex behavior patterns to calculate the necessary wing movements in order to follow the desired trajectory optimally.

The flying fox receives the necessary control algorithms from the control computer, where they are learned by machine and permanently improved. This allows the BionicFlyingFox to optimize its behavior during flights and thus to fly more precisely along the specified trajectories from lap to lap. Control is effected via the movement of the legs and the thus changeable wing area.