xbpeng/MimicKit
A lightweight suite of motion imitation methods for training controllers.
What it solves
MimicKit is a lightweight framework designed to train motion controllers for characters (such as humanoids or quadruped robots) so they can imitate specific reference motions. It provides a standardized way to implement and test various motion imitation methods using different physics simulators.
Related
- Project
zitongbai/legged_labA reinforcement learning extension for Isaac Lab that enables humanoid robots, such as the Unitree G1, to learn complex motions via DeepMimic and Adversarial Motion Priors.
- Project
NVlabs/ProtoMotionsProtoMotions 3 is an Apache‑2.0‑licensed, GPU‑accelerated framework for large‑scale motion learning, retargeting, and sim‑to‑real deployment of humanoid robots. It supports multi‑GPU training on datasets such as AMASS and BONES, one‑command retargeting via PyRoki, cross‑simulator testing (Isaac Gym, Newton, MuJoCo), and export of ONNX policies for zero‑shot deployment on real hardware (e.g., Unitree G1).
- Project
ARISE-Initiative/robomimicA framework for robot learning from demonstration that provides standardized datasets and offline learning algorithms to make robot manipulation research accessible and reproducible.
- Project
lenskit/lkpyLensKit for Python is a toolkit for researchers and students to build, run, and evaluate recommender system algorithms in a flexible, experimental environment.
- Project
microsoft/agent-lightningA lightweight agentic RL framework that allows developers to train AI agents with reinforcement learning using their existing harnesses and environments with minimal code changes.