Bionic Learning Network
Inspired by nature
Nature never ceases to amaze us with its aesthetic beauty. It is no surprise therefore that nature’s fascinating creatures also served as inspiration for the new projects of Festo’s Bionic Learning Network. The structure and kinematics of AquaJelly and AirJelly are based on their biological model, the jellyfish. The test beds utilize cutting-edge technologies. AquaJelly, for example, is a man-made autonomous jelly fish with an electric drive and an intelligent, adaptive mechanical system. Festo’s YoYo product illustrates that fully automatic control is possible with the aid of mechatronics – it also forms a bridge to the company’s core competency, automation with air.
AquaJelly is an artificial autonomous jellyfish with an electric drive and an intelligent, adaptive mechanical system. AquaJelly consists of a translucent hemisphere and eight tentacles used for propulsion. At the centre of the AquaJelly is a watertight, laser-sintered pressure vessel. This comprises a central, electric drive, two lithium-ion-polymer batteries, the charge control device and the servo motors for the swashplate.
The structure of each tentacle uses the Fin Ray Effect® – a construction design derived from the functional anatomy of a fish fin. It moves with the aid of a peristaltic propulsion system, or wave-like contractions, based on the reaction thrust principle used by its biological precursor. The motion of the AquaJelly in three-dimensional environments is controlled by shifting its weight. Two servo motors integrated into the central pressure vessel actuate a swashplate. This swashplate controls a four-arm pendulum which can be steered in the four spatial directions. When a pendulum moves in a certain direction, the centre of gravity of the jellyfish changes in this direction.
AquaJelly is capable of independently controlling its own energy supply, by means of communication between the AquaJelly and a charging station. Whenever the AquaJelly comes to a charger located above the water basin, it is sucked towards it and provided with electricity.
For communication on the water surface, the AquaJelly can use the energy-conserving short-range radio standard ZigBee, which enables it to exchange status details with the charger and signal to other AquaJellies on the surface that the charger is occupied.
The main communication medium under water, however, is light. The AquaJelly has eleven infrared light-emitting diodes with which it can communicate over distances of up to approx. 80 cm. The pulsed infrared signals are sent from inside an almost spherical structure around the AquaJelly. On receiving a position signal from an approaching jellyfish, for example, the AquaJelly can start its evasion manoeuvre in plenty of time. In addition to environment sensors, the AquaJelly also has internal sensors which monitor its energy level, as well as a pressure sensor which allows it to gauge its depth in the basin to within a few millimetres.
Each jellyfish decides autonomously which action to carry out on the basis of its current condition. This central electric drive, combined with an adaptive mechanical system and intelligent autonomous electronics, opens up possible new applications for self-controlling systems. If a large number of AquaJellies were equipped with communicative abilities, these could act like a shoal with the behaviour pattern of a more highly developed system. If one applies this principle to automation, then numerous autonomous or semi-autonomous intelligent systems might be able to work together. In this way, large problems could be solved by small systems working together in harmony.
Air is the element of the AirJelly. Rather than swimming through water like the AquaJelly, it glides through the air with the aid of its central electric drive and an intelligent, adaptive mechanical system. The remote-controlled AirJelly is kept in the air by its helium-filled ballonet.
The AirJelly’s only energy source are two lithium-ion-polymer batteries, to which the central electric drive is attached. This transmits its power to a bevel gear and then to eight spur gears, which drive the eight tentacles of the jellyfish via their respective cranks. The structure of each tentacle is based on the Fin Ray Effect®. Using a peristaltic movement to drive a balloon was previously unknown in the history of aviation. The AirJelly is the first indoor flying object to use such a peristaltic propulsion system. The jellyfish glides gently through the air thanks to this new drive concept based on the reaction thrust principle.
The AirJelly steers through three-dimensional environments by shifting its weight. Its two servo motors are located at the “North pole” of the jellyfish and controlled proportionally. If the pendulum moves in one direction, the AirJelly’s centre of gravity shifts in this direction – the AirJelly is thus able to swim in any spatial direction. The propulsive force of the drive can be varied by moving the Fin Ray® tentacles more quickly or slowly.
Festo demonstrates with this exhibit that a central electric drive – combined with an intelligent mechanical system – can offer fascinating possibilities for “lighter-than-air” aviation. Festo aims to delight its customers with innovative, fascinating and intelligent solutions in both automation and didactics. It therefore offers a wide range of electric, pneumatic and hybrid drive systems, together with the respective sensors and control possibilities.
The biological inspiration for the AirArm was the result of an analysis of lobster and locust legs as well as of the human pointing movement. The two-part exoskeleton arm is powered by pneumatic muscles. With the aid of intelligent control technology, the AirArm can even catch water drops.
The technical purpose of the AirArm is to be able to reach as many points as possible from one position in a hemispherical working space. The general technical principle chosen was a two-part folding system with counteracting muscles as drive pairs.
In order to combine lightness with rigidity, the triangular recesses of the arm modules were modelled on the exoskeleton of a locust’s leg. The cross-staggered articulated axles of the lobster leg in combination with adapted segment lengths are sufficiently flexible for reaching movements in a hemispherical work area. By using pneumatic muscles as its drives, the AirArm can achieve a high degree of resilience when holding a certain position, while at the same time consuming minimal energy.
The pneumatic muscles used for the drive boast a very favourable ratio between the high mechanical forces they generate and their low weight. The AirArm is therefore well suited to fast dynamic movements. However, the pneumatic muscles demonstrate high elasticity and non-linear relationships between travel, pneumatic pressure and generated force. Just like its biologic role model, the technical system has to learn to cope with this – the task of the AirArm’s adaptive control system!
The entire control system is designed on the basis of a model. The mathematical model of the AirArm is first generated by the computer, which is then used to design and optimize the control system. The entire control system is translated into a program using automatic code generation and then transferred to an industrial PC which controls the AirArm in real time.
Playing with yo-yos requires great patience and skill. With its own particular YoYo, Festo aims to demonstrate that control systems can also be fully automated with the aid of mechatronics. Festo has thus formed a bridge to its core competency, automation with air.
The YoYo consists of three independent and different sized YoYos of 16", 20" and 24". The YoYos are all designed as various types of Maxwell wheels and each is driven by a Fluidic Muscle.
Each muscle is mounted vertically and fixed at its upper end, where it is supplied with air. Hanging at the lower contracting end is the Maxwell wheel held by two tear-resistant cords. The exact vertical position of the Maxwell disk is measured by an ultrasonic sensor mounted in a perpendicular position under the wheel. Once the wheel has reached the right point for the force input, a signal is generated to open the switching valve. The compressed air released flows into the muscle, which suddenly contracts and pulls the wheel up via the tightened cords. At this point, energy is fed into the system. After a brief moment, the valve is closed again and the air in the muscle is slowly released. The muscle expands gently into its original position and reduces the cord tension between muscle and wheel. Just like a hand-held yo-yo, a hanging cord supports the rollup motion of the rotating disk. At bottom dead centre, the muscle has regained its original length and the YoYo at least its starting height.
In order to play the yo-yo, energy has to be provided at the right moment and in the right amount. This overall mechatronic design is enabled by combining a control system with intelligent sensor technology. Due to its fast and precise delivery of tractive force, Festo’s Fluidic Muscle is ideally suited for the drive task. The YoYo helps show off its benefits as a pure traction actuator.
Bionic Learning Network
The Bionic Learning Network is part of Festo’s commitment to technical education and training. In cooperation with students, major-name universities, institutes and development companies, Festo supports projects and prototypes which extend beyond its core segments of Automation and Didactic and which may produce interesting application fields of the future. The aim is to make automated movements even more efficient and productive with the aid of bionics. The Bionic Learning Network demonstrates fascinating solutions to complex problems.
Fin Ray Effect® is a trademark of Evologics GmbH.
The press text and photos are available on the Internet at www.festo.com/press,
for further information on Festo's Bionic Learning Network please refer to www.festo.com/bionic
Please refer to: Festo press photos
Please refer to: Festo press photos
Please refer to: Festo press photos
Please refer to:
Festo press photos