Robotic hand mimics human hand : next software brain control
The European project called Sensopac, whose goal is to create a robotic arm, hand and brain with human-like physical and cognitive capabilities, is made up of 12 groups and has now made advances with its robot hand. The hand mimics the flexibility and sensitivity of a human hand and is controlled by a neural-network-based program modelled on the cerebellum. In the next six months Edinburgh and Lund Universities will see how well their system can control arm and hand.
Scientists at the German Aerospace Centre (DLR) guided by physiology results from Umeå University, in Sweden, 
made a robotic "skin" out of a thin, flexible carbon that changes its resistance depending on pressure. This allows the robot hand to tell the shapes of an object, the amount of force placed upon it, and the direction of that force. This lets them combine information from sensors in different parts of the skin in order to minimise the number of information-carrying wires to five. Right Sensopac project Courtesy: http://www.blogsforcompanies.com
The artificial arm that DLR built and are now experimenting with uses a total of 58 motors in opposing pairs, coupled with non-linear springs, to control the arm.
Thirty-eight opposing motors control the hand's joints, giving it a touch that ranges from light to forceful. The researchers modeled the robot hand by using hundreds of MRI images of human hands.
As for the robot's learning ability, the team hopes to improve its understanding of movement and sensation through its neural network. When the robot picks up a cup, it will be able to sense the properties within and adjust its motions depending on whether the cup contains water or flour, for example.
Scientists at the University of Edinburgh, in Scotland, and at Lund University, in Sweden, decided that the best approach was to model the human cerebellum. The cerebellum is a fist-sized organ at the base of the brain that coordinates sensation and movement. The researchers are currently using software to simulate important aspects of how the cerebellum processes and integrates information.
“It’s the first neural-network-based controller that can control the dynamics of a robotic system in its full operational range,” says Patrick van der Smagt, the coordinator of Sensopac. In the next six months, they will be seeing how well this system can learn to control the arm.
Source: http://cordis.europa.eu/ictresults/