Be notified of new releases
Create your free GitHub account today to subscribe to this repository for new releases and build software alongside 28 million developers.Sign up
Bullet 2.87 has improved support for robotics, reinforcement learning and VR. In particular, see the "Reinforcement Learning" section in the pybullet quickstart guide at http://pybullet.org . There are also preliminary C# bindings to allow the use of pybullet inside Unity 3D for robotics and reinforcement learning.
Here are some videos:
- Minitaur pybullet trained with TF Agents PPO, improved motor model, randomization
- pybullet Ant trained using TensorFlow Agents
- pybullet KUKA grasp training/enjoy using Tensorflow+OpenAI gym+baselines DQN
Some example training the pybullet_pendulum using TensorFlow Agents PPO:
pip install pybullet, agents, tensorflow, gym
python -m pybullet_envs.agents.train_ppo --config=pybullet_pendulum --logdir=pendulum
tensorboard --logdir=pendulum --port=2222
python -m pybullet_envs.agents.visualize_ppo --logdir=pendulum/xxxxx --outdir=pendulum_video
The Bullet 2.86 has improved Python bindings, pybullet, for robotics, machine learning and VR, see the pybullet quickstart guide.
Furthermore, the PGS LCP constraint solver has a new option to terminate as soon as the residual (error) is below a specified tolerance (instead of terminating after a fixed number of iterations). There are optional multithreaded optimizations, thanks to lunkhound. There is preliminary support to load some MuJoCo MJCF xml files (see data/mjcf),see Bullet VR haptic experiments with a VR glove:
Bullet 2.85 (previously known as 2.84) introduces pybullet, easy to use Python bindings, as well as Virtual Reality support for HTC Vice and Oculus Rift. In addition, there is support for Inverse Kinematics and Inverse Dynamics. This release is marked as 'prerelease' until the documentation is updated. See also this video: https://www.youtube.com/watch?v=VMJyZtHQL50