MyoDex: Generalizable Representations for Dexterous Physiological Manipulation

The complexity of human dexterity has attracted attention from multiple fields. Still, much is to be understood about how hand manipulation behaviors emerge. In this work we aim at learning dexterous manipulation behaviors with a physiologically realistic hand model: MyoHand. In contrast to prior works demonstrating isolated postural and force control, here we demonstrate musculoskeletal agents (MyoDex) exhibiting contact-rich dynamic dexterous manipulation behaviors in simulation. Furthermore, to demonstrate generalization, we show that a single MyoDex agent can be trained to solve up-to 14 different contact-rich tasks. Aligned with human development, simultaneous learning of multiple tasks imparts physiological coordinated muscle contractions i.e., muscle synergies, that are not only shared amongst those in-domain tasks but are also effective in out-of-domain tasks. By leveraging these pre-trained manipulation synergies, we show generalization to 14 additional previously unsolved tasks. While physiological behaviors with large muscle groups (such as legged-locomotion, arm-reaching, etc), have been demonstrated before, to the best of our knowledge nimble behaviors of this complexity with smaller muscle groups are being demonstrated for the first time.