# rohanpsingh/LearningHumanoidWalking

Training a humanoid robot for locomotion using Reinforcement Learning

Repository: https://github.com/rohanpsingh/LearningHumanoidWalking
Canonical: https://ross.abutalabs.com/products/learninghumanoidwalking
Language: Python
License: BSD-2-Clause
License Family: permissive
Topics: bipedal-robots, cassie, humanoids, ppo, reinforcement-learning, jvrc-1, mujoco
Last push: 2026-05-03T01:05:26+00:00

## Health v2 (maintenance only)
Score: 70/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 80, release rhythm 40, longevity 100
- inputs: {"age_days": 1512, "days_push": 123, "days_rel": 554, "gap_med": 6, "n_releases_24m": 2}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1207, forks 135 (observed 2026-08-28T04:03:59.513805+00:00)

## What it is
A Python research codebase for training humanoid robots to walk using deep reinforcement learning (PPO) in MuJoCo simulation. It provides extensible environments, task/reward definitions, and robot abstractions for models like JVRC-1 and Unitree H1, accompanying peer-reviewed papers on robust bipedal locomotion.

## Use cases
- train a humanoid robot to walk with reinforcement learning
- simulate bipedal locomotion in mujoco
- learn ppo policies for humanoid walking
- add a custom humanoid robot to rl locomotion training
- reproduce humanoid walking research papers
- train walking policies for jvrc-1 or unitree h1

## When to choose
- You want to train bipedal or humanoid locomotion policies with PPO in MuJoCo
- You need a reference implementation tied to published humanoid walking research
- You want a structured, extensible codebase for adding your own humanoid robot environment

## When to avoid
- You need a general-purpose RL library with many algorithms beyond PPO
- You are working on quadrupeds, manipulators, or non-humanoid robot locomotion
- You need a production robotics middleware or real-time control stack rather than training code

## Facets
- artifact type: library
- maturity: active
- function: reinforcement-learning, simulation, machine-learning
- domain: robotics, reinforcement-learning, simulation, machine-learning
- platform: python, windows
- tags: humanoid-robots, bipedal-locomotion, mujoco, ppo, jvrc-1, unitree-h1, sim-to-real, research-code, physics-simulation, robot-controllers, linux, macos

## Member repositories
- rohanpsingh/LearningHumanoidWalking (main) score 70

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:59.513805+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 2026-08-30T06:19:14.774742+00:00, confidence not recorded.
  - readme: https://github.com/rohanpsingh/LearningHumanoidWalking (fetched 2026-08-28T04:03:59.513805+00:00, sha b810407d606a)
- Data as of 2026-08-30T08:39:29.467469+00:00.