Communications of the ACM
Robotic Elephant Trunk Can Learn Tasks on Its Own
Researchers at the University of Tubingen in Germany used three-dimensional printing to create a robotic elephant trunk that utilizes artificial intelligence to emulate how sensory input triggers synaptic chain reactions in organic brains.
Credit: New Scientist
Researchers at Germany's University of Tubingen used three-dimensional (3D) printing to create a robotic elephant trunk that utilizes artificial intelligence (AI) to emulate how sensory input triggers synaptic chain reactions in organic brains.
Tubingen's Sebastian Otte and colleagues assembled the device from modules with gear-driving motors that tilt up to 40 degrees in two axes.
The AI was trained on examples of the motor inputs required to move the trunk in certain ways, and tests showed it could direct the tip of the trunk within less than a centimeter from a target.
The robot is a proof of concept of a spiking neural network algorithm, which works like an actual brain in which certain inputs cause a chain reaction of firing synapses, while requiring orders of magnitude less computational power and energy.
From New Scientist
View Full Article
Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA
No entries found