Page 13: Research news on Embodied robotic manipulation

Embodied robotic manipulation investigates robotic and prosthetic limbs that physically interact with the environment using human-like, adaptive control. Work in this area integrates soft robotic structures, tendon-driven and biohybrid actuators, and exoskeletons with rich multimodal sensing, including vision, tactile, and proprioceptive feedback. Machine learning methods such as imitation learning, reinforcement learning, and meta-learning are used to acquire dexterous skills, enable shared and autonomous control, and support intuitive human–robot interaction through haptic interfaces, brain–computer interfaces, and teleoperation systems.

Machine learning & AI

Energy and memory: A new neural network paradigm

Listen to the first notes of an old, beloved song. Can you name that tune? If you can, congratulations—it's a triumph of your associative memory, in which one piece of information (the first few notes) triggers the memory ...

Engineering

'Countersnapping' structures shrink when pulled

When you pull something—like a rubber band—you expect it to get longer. But what if it did the opposite? What if it suddenly shrank instead? In a study published in Proceedings of the National Academy of Sciences, researchers ...

page 13 from 16