LearningGripper – gripping and positioning through independent learning

The LearningGripper from Festo looks like an abstract form of the human hand. The four fingers of the gripper are driven by twelve pneumatic bellows actuators with low-level pressurisation. Thanks to the machine learning process, the gripper is able to teach itself to carry out complex actions such as the targeted gripping and positioning of an object.

Loading the player ...

 

Reduced programming effort thanks to machine learning

The LearningGripper trade fair display demonstrates a gripper which takes less than an hour to learn a mechanical motion strategy – from its first attempt to the reliable execution of the required task. And the gripper is only told what to do, but not how to do it.   The complex strategy required for this is developed independently by the gripper’s learning algorithms – without any further programming.

 

A second gripper demonstrates a process it had previously learned within the desired target scenario: it lifts the ball and turns it so that the embossed lettering can finally be seen at the top

LearningGripper: self-learning gripper system for positioning round objects

LearningGripper: self-learning gripper system for positioning round objects

Self-learning: The LearningGripper’s four pneumatic fingers …

Self-learning: The LearningGripper’s four pneumatic fingers …

… position the ball until the correct side is at the top.

… position the ball until the correct side is at the top.

Smart: The learning algorithms replace complex programming …

Smart: The learning algorithms replace complex programming …

… and allow for quick commissioning of the system.

… and allow for quick commissioning of the system.