4/5/2013

Festo presents current projects from the Bionic Learning Network

Festo presents current projects from the Bionic Learning Network

Inspired by nature
Each year, Festo’s Bionic Learning Network provides new inspiration for automation technology at the Hanover Trade Fair.

Festo presents current projects from the Bionic Learning Network

Inspired by nature

 

Each year, Festo’s Bionic Learning Network provides new inspiration for automation technology at the Hanover Trade Fair – and this year is no different. The emerging fields include research in the areas of function integration, lightweight construction, self-configuration and machine learning. With the “BionicOpter”, the “WaveHandling” system and the “LearningGripper”, Festo shows how principles from nature can be applied in automation technology.

 

As a global manufacturer of pneumatic and electric automation technology, Festo’s core business is helping to shape the production and working environments of the future and offering its customers innovative solutions for the production systems of tomorrow and beyond. “This is essential for our long-term reputation as a competent partner with a high level of problem-solving skills,” emphasises Dr.-Ing. Heinrich Frontzek, Head of Corporate Communication and Future Concepts. “What we need to do is simplify the challenges involved in production sequences and guarantee intuitive control of machines and plants. The current projects from Festo’s Bionic Learning Network provide visonary approaches on how to do this,” says Mr. Frontzek.

 

Inspired by dragonfly flight

After bird flight had been deciphered with the SmartBird in 2011, the developers took on their next-biggest challenge in the Bionic Learning Network: modelling the dragonfly at a technical level. The BionicOpter is an ultralight flying object. Just like its model in nature, the BionicOpter can fly in all directions and execute the most complicated flight manoeuvres. The BionicOpter’s ability to move each of its wings independently enables it to slow down and turn abruptly, to accelerate swiftly and even to fly backwards. This means that for the first time there is a model that can master all the flight conditions of a helicopter, plane and even a glider. Despite its complexity, the highly integrated system can be operated easily and intuitively via a smartphone.

 

This unique way of flying is made possible by lightweight construction and the integration of functions: components such as sensors, actuators and mechanical components as well as open- and closed-loop control systems are installed in a very tight space and adapted to one another.

The flapping frequency, amplitude and angle of incidence are controlled by software and electronics; the pilot just has to steer the dragonfly – there is no need to coordinate the complex motion sequences.

 

The principles of ultra-lightweight construction are applied throughout the flying object. With a wingspan of 63 cm and a body length of 44 cm, the model dragonfly weighs just 175 grams. The wings consist of a carbon-fibre frame and a thin foil covering. The intelligent kinematics correct any vibrations during flight and ensure flight stability. In order to stabilise the flying object, data on the position and the twisting of the wings is continuously recorded and evaluated in real time during the flight of the dragonfly.

 

Modular conveyor: conveying and sorting in one

With the WaveHandling pneumatic conveyor, engineers from Festo’s Bionic Learning Network have developed a modular system that can move a surface in such a way that objects are transported and sorted purposefully. Thanks to the integration of a sorting function, an additional handling unit is no longer required for this process. The conveyor consists of numerous bellows modules that deform the surface creating a wave motion that transports the objects in a targeted manner.

 

Inspiration for this principle was provided by waves in nature. The movement of the wind over the smooth surface of the water produces small ripples, which grow as the wind pushes against them. However, what is being moved by the waves is energy, not water. The water molecules within a wave move up and down in a circular motion, but remain in roughly the same place. Yet the energy produced causes the wave to roll over the surface of the sea. The WaveHandling system behaves in a similar way: while each individual bellows advances and retracts in the same spot, a wave moves over the surface of the conveyor.

 

The individual modules are self-configuring. This means that the system can be started up quickly and without programming, no matter what the layout is. A potential application of the platform is in the food industry, for automatically transporting delicate items like fruit and vegetables and sorting them for the next process step. With the WaveHandling transport system, Festo is already demonstrating how the configuration of a system will be handled by the individual modules themselves in the future.

 

Self-learning gripper: LearningGripper

The LearningGripper is a four-fingered gripper that looks like an abstract form of the human hand. The four fingers of the gripper are actuated pneumatically by 12 bellows actuators with low-level pressurisation. What makes this bionic gripper so special is its learning ability. This is thanks to learning algorithms which take the place of highly complex programming.

 

Thanks to machine learning, an offshoot of artificial intelligence, the gripper can teach itself to carry out complex tasks such as gripping and positioning a ball. In concrete terms, the gripper assigns itself the task of turning a ball so that a particular point of the ball points upwards. It picks up the corresponding motion sequences through learning by reinforcement.

 

With this principle, self-learning systems like the LearningGripper could be built into future production lines and autonomously optimise their own performance.

 

Festo at a glance

Festo is a global leader in automation technology, and the world market leader in industrial training and education. Festo’s pneumatic and electric drive technology stands for innovation in industrial and process automation – from individual products to ready-to-install solutions. Innovation for the best possible productivity of our customers, global presence and close, long-term partnerships with our customers are the hallmarks of Festo. 16,200 employees work worldwide in 61 national companies of Festo. Festo’s turnover for 2012 was 2.2 billion euros.

www.festo.com

 

About the Bionic Learning Network

The Bionic Learning Network is a network linking Festo to well-known universities, institutes and development companies. The objective of this initiative is to transfer biological principles to industrial technology and to produce innovative solutions and visions for industrial applications, all through bionics. Automated motion sequences can be made even more energy-efficient and productive using bionics, potentially providing industry with completely new solutions for practical problems.

 

You will find Future Concepts exhibits from Festo at stand D07 in hall 15.

Further information can be found at: www.festo.com/bionics

Full article with pictures: www.festo.com/press 

 

 

For queries, please contact:

Festo AG & Co. KG

Simone Schmid, Corporate Communication – Technology

Tel.: +49 (0) 711-347-57489, E-mail: smns@de.festo.com

Contact:

Simone Schmid

+49(711)347-57489


Festo Media Service

 
 

Photos

Caption Download
BionicOpter BionicOpter – inspired by dragonfly flight
WaveHandling WaveHandling – transporting and sorting in one
LearningGripper LearningGripper – gripping and positioning through machine learning
BionicOpter The BionicOpter is an ultralight flying object.
BionicOpter The highly integrated system can be operated easily and intuitively via a smartphone.
BionicOpter Just like its model in nature, the BionicOpter can fly in all directions and execute the most complicated flight manoeuvres.

Videos

Description Download
WaveHandling Conveying and sorting in one
2,31 Min
LearningGripper Gripping and positioning through independent learning
1,32 Min
BionicOpter Inspiration dragonfly flight
1,27 Min
BionicOpter Animation
1,25 Min
BionicOpter Animation
1,25 Min