Meet the BionicSoftHand: a pneumatic robot hand with AI and deep learning that learns to grasp different objects and turn them. It is modeled after the human hand and uses reinforcement learning to complete a wide variety of tasks.
This robotic hand features a flexible printed circuit board, inertial sensors, tactile force sensors, and elastomer bellows with two air chambers for movement. As Festo researchers explain:
the BionicSoftHand has no bones. It controls its movements via the pneumatic bellows structures in its gripper fingers. When the chambers are filled with air, the gripper fingers bend. If the air chambers are empty, the gripper fingers remain stretched. The thumb and index finger are additionally equipped with a swivel module, which allows these two gripper fingers to be moved laterally. This gives the bionic robot hand a total of twelve degrees of freedom.