osim-rl package allows you to synthesize physiologically accurate movement by combining biomechanical expertise embeded in OpenSim simulation software with state-of-the-art control strategies using Deep Reinforcement Learning.

HUMAN environment

Our objectives are to:

  • use Reinforcement Learning (RL) to solve problems in healthcare,
  • promote open-source tools in RL research (the physics simulator, the RL environment, and the competition platform on which we run challenges are all open-source),
  • encourage RL research in computationally complex environments, with stochasticity and highly-dimensional action spaces, relevant to real-life applications,
  • bridge biomechanics, neuroscience, and computer science communities.