body model

December 30, 2011 - Lingchu Intelligence released the first end-to-end Reinforcement Learning (RL)-based embodied model, Psi R0. 1AI has learned that the model supports dual dexterous hands to collaborate on complex operations, mixing multiple skills in tandem and generating intelligences with reasoning capabilities to complete and close the loop on long-range dexterous operation tasks. Moreover, Psi R0 can also realize cross-item and cross-scene level generalization. Take e-commerce scenarios as an example, product packing is a typical long-range task that requires tens of thousands of products to be grabbed, scanned, placed, and tied in a plastic bag, etc. Psi R0 is able to ...