Science Daily January 30, 2019
Researchers at Columbia University have created a robot that learns what it is, from scratch, with zero prior knowledge of physics, geometry, or motor dynamics. After a brief period of “babbling,” and within about a day of intensive computing, the robot creates a self-simulation and uses that self-simulator internally to contemplate and adapt to different situations, handling new tasks as well as detecting and repairing damage in its own body. Using a four-degree-of-freedom articulated robotic arm, initially the robot moved randomly and collected approximately one thousand trajectories. Then used deep learning to create a self-model. After less than 35 hours of training, the self-model became consistent with the physical robot to within about four centimeters…read more. TECHNICAL ARTICLEÂ
An image of the deformed robotic arm in multiple poses as it was collecting data through random motion. Credit: Robert Kwiatkowski/Columbia Engineering