This article is about recognizing handheld objects from incomplete tactile observations with a classifier trained on only visual representations. Our method is based on the deep learning (DL) architecture PointNet and a curriculum learning (CL) technique for fostering the learning of descriptors robust to partial representations of objects. The learning procedure gradually … [Read more...] about Visuo-Tactile Recognition of Partial Point Clouds Using PointNet and Curriculum Learning
Shape control
Tensegristats: Lightweight Untethered Morphing Robots
Flexible and shape-shifting structures are common in nature. Creating shape-shifting machines using rigid components requires complex design work. Soft robotic attempts are closer to the muscular hydrostats they mimic but are slow and inefficient untethered and do not generally allow large-scale shape change between forms. Here we present the idea of “tensegristat” robots in … [Read more...] about Tensegristats: Lightweight Untethered Morphing Robots
