# TeleHuman/PBHC

Official Implementation of "KungfuBot: Physics-Based Humanoid Whole-Body Control for Learning Highly-Dynamic Skills"

Repository: https://github.com/TeleHuman/PBHC
Canonical: https://ross.abutalabs.com/products/pbhc
Homepage: https://kungfubot.github.io/
Language: Python
License Family: other
Last push: 2026-07-01T06:40:40+00:00

## Health v2 (maintenance only)
Score: 59/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 90, release rhythm 35, longevity 33
- inputs: {"age_days": 470, "days_push": 63, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1055, forks 141 (observed 2026-08-28T04:03:24.345411+00:00)

## What it is
Official implementation of KungfuBot, a physics-based humanoid whole-body control framework for learning highly-dynamic skills like kungfu and dancing via reinforcement learning. It includes motion processing pipelines and adaptive motion tracking, deployed on the Unitree G1 robot.

## Use cases
- train humanoid whole-body control policies for dynamic motions
- imitate kungfu and dance movements on a humanoid robot
- retarget and filter human motion capture data for robots
- deploy RL policies on Unitree G1
- research physics-based motion tracking with adaptive curricula

## When to choose
- you need a humanoid controller for highly-dynamic, aggressive motions
- you want a research baseline for physics-based motion imitation
- you deploy on Unitree G1 hardware

## When to avoid
- you need a general-purpose robot middleware or ROS stack
- you want a production-ready, licensed product (no license is provided)
- your robot is not a humanoid or lacks Isaac-based simulation support

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, reinforcement-learning, simulation, robotics
- domain: robotics, reinforcement-learning, machine-learning, simulation
- platform: python
- tags: humanoid-robotics, motion-imitation, whole-body-control, unitree-g1, physics-based-control, research-code, neurips-2025, linux, gpu

## Member repositories
- TeleHuman/PBHC (main) score 59

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:24.345411+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:58:28.489858+00:00, confidence not recorded.
  - readme: https://github.com/TeleHuman/PBHC (fetched 2026-08-28T04:03:24.345411+00:00, sha bde5c3609f8e)
  - homepage: https://kungfubot.github.io/ (fetched 2026-08-29T13:00:20.749889+00:00, sha f22157a7b7d2)
- Data as of 2026-08-30T08:39:29.467469+00:00.