Playing with MuJoCo is a lot of fun. I recommend it for people who just want to experiment with RL. You can do random weird stuff like this: https://www.youtube.com/watch?v=e-9u_vepfY8. I'm really glad DeepMind is continuing to update it.
Holy Molly. The live demo where you put a place holder and the trained robot dog just goes there is insane! You can also apply force (aka kick) the robot while it is moving!
An open-source library for GPU-accelerated robot learning and sim-to-real transfer.
Really cool. If we could do sim training to rl conversion in robotics, we migth get a llm moment for robotics, with good pretrained movement algorithms.
This is great - I haven't looked at the documentation yet but do you support soft body simulation?