AI use cases
Simulation-trained reinforcement-learning policies learn goal-conditioned hand and arm control that can generalize zero shot to unseen tools and tasks.
From Why Robotics Still Isn't Solved - But Could Be Soon | YC Paper Club by Y Combinator