This website uses cookies primarily for visitor analytics. Certain pages will ask you to fill in contact details to receive additional information. On these pages you have the option of having the site log your details for future visits. Indicating you want the site to remember your details will place a cookie on your device. To view our full cookie policy, please click here. You can also view it at any time by going to our Contact Us page.

BionicFlyingFox uses machine learning to determine ideal flight path

25 April 2018

Festo’s BionicFlyingFox features on-board electronics with an external motion-tracking system to help it move semi-autonomously in a defined airspace.

To emulate the natural flying fox as closely as possible, the wing kinematics of the BionicFlyingFox are divided into primaries and secondaries; all the joints lie in the same plane. The wings are covered with an elastic membrane, which extends down to the feet. This flying membrane is wafer-thin and ultralight, but also robust.

Motion tracking system for semi-autonomous flight

To enable the BionicFlyingFox to move semi-autonomously within a defined space, it communicates with a so-called motion tracking system. The installation with two infrared cameras constantly records its position. The cameras, which are mounted to a pan-tilt unit, can be rotated and inclined in such a way as to track the entire flight of the BionicFlyingFox from the ground. At the same time, the motion tracking system plans the flight paths and issues the required control commands. Starting and landing are performed by the human operator; the autopilot takes over in flight.

Machine learning for the ideal flight path

The images from the cameras are conveyed to a central master computer, which evaluates the data and externally coordinates the flight like an air traffic controller. Pre-programmed flight routes stored on the computer specify the path taken by the BionicFlyingFox in performing its manoeuvres. The wing movements required to ideally implement the intended movement sequences are calculated by the artificial flying fox itself with the help of its on-board electronics and its complex behaviour patterns. The flying fox receives the necessary control algorithms from the master computer, where they are automatically learnt and constantly improved. The BionicFlyingFox is thus able to optimise its behaviour during flight and thereby follow the specified courses more precisely with each circuit flown.

Innovative flying membrane for various different applications

The innovative flying membrane was specially developed by the bionics team for the BionicFlyingFox. It consists of two airtight foils and a woven elastane fabric, which are welded together at approximately 45,000 points. The fabric’s honeycomb structure prevents small cracks in the flying membrane from increasing in size. The BionicFlyingFox can thus continue flying even if the fabric sustains minor damage. Due to its elasticity, the flying membrane stays almost uncreased even when the wings are retracted. Since the foil is not only elastic, but also airtight and lightweight, it could also potentially be used in other flying objects or for clothing design and in the field of architecture.

To take a closer look, watch this video.


Contact Details and Archive...

Print this page | E-mail this page