Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Training a humanoid robot for locomotion using Reinforcement Learning
| Date | Stars |
|---|---|
| 2026-07-24 | 1193 |
| 2026-07-25 | 1193 |
| 2026-07-28 | 1193 |
| 2026-07-30 | 1193 |
| 2026-07-31 | 1200 |
| 2026-08-02 | 1202 |
| 2026-08-03 | 1203 |
| 2026-08-04 | 1201 |
| 2026-08-05 | 1203 |
| 2026-08-06 | 1203 |
Today
— stars today
This week
+10 stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.84%/day
# LearningHumanoidWalking <p align="center"> <a href="https://www.youtube.com/watch?v=ZgfNzGAkk2Q"><img src="https://github.com/user-attachments/assets/5211cdcd-2267-497b-bd66-ac833703a134" alt="humanoid-walk" style="width:1000px"/></a> </p> Code for the papers: - [**Robust Humanoid Walking on Compliant and Uneven Terrain with Deep Reinforcement Learning**](https://ieeexplore.ieee.org/abstract/document/10769793) [Rohan P. Singh](https://rohanpsingh.github.io), [Mitsuharu Morisawa](https://unit.aist.go.jp/jrl-22022/en/members/member-morisawa.html), [Mehdi Benallegue](https://unit.aist.go.jp/jrl-22022/en/members/member-benalleguem.html), [Zhaoming Xie](https://zhaomingxie.github.io/), [Fumio Kanehiro](https://unit.aist.go.jp/jrl-22022/en/members/member-kanehiro.html) - [**Learning Bipedal Walking for Humanoids with Current Feedback**](https://arxiv.org/pdf/2303.03724.pdf) [Rohan P. Singh](https://rohanpsingh.github.io), [Zhaoming Xie](https://zhaomingxie.github.io/), [Pierre Gergondet](https://unit.aist.go.jp/jrl-22022/en/members/member-gergondet.html), [Fumio Kanehiro](https://unit.aist.go.jp/jrl-22022/en/members/member-kanehiro.html) - [**Learning Bipedal Walking On Planned Footsteps For Humanoid Robots**](https://arxiv.org/pdf/2207.12644.pdf) [Rohan P. Singh](https://rohanpsingh.github.io), [Mehdi Benallegue](https://unit.aist.go.jp/jrl-22022/en/members/member-benalleguem.html), [Mitsuharu Morisawa](https://unit.aist.go.jp/jrl-22022/en/members/member-morisawa.html), [Rafael Cisneros](https://unit.aist.go.jp/jrl-22022/en/members/member-cisneros.html), [Fumio Kanehiro](https://unit.aist.go.jp/jrl-22022/en/members/member-kanehiro.html) ## Code structure: A rough outline for the repository that might be useful for adding your own robot: ``` LearningHumanoidWalking/ ├── envs/ <-- Environment implementations │ ├── common/ │ │ ├── base_humanoid_env.py <-- Base class for all humanoid environments │ │ ├── mujoco_env.py <-- MuJoCo simulation wrapper │ │ └── robot_interface.py <-- Robot state/control abstraction │ ├── jvrc/ <-- JVRC robot environments │ └── h1/ <-- Unitree H1 robot environment ├── tasks/ <-- Task definitions (rewards, termination) ├── rl/ <-- Reinforcement learning ├── robots/ <-- Robot abstractions (PD control, stepping logic) ├── models/ <-- MuJoCo model files └── tests/ <-- Test suite ``` ### Key abstractions: - **BaseHumanoidEnv**: Common functionality for humanoid environments (observation history, action smoothing, reset logic) - **BaseTask**: Interface for task implementations (reset, step, calc_reward, done) - **Reward functions**: Explicit parameter functions in `tasks/rewards.py` for testability ## Requirements: - Python version: >= 3.10 - [uv](https://docs.astral.sh/uv/) package manager ```bash $ uv sync ``` ## Usage: Environment names supported: | Task Description | Environment name | | ----------- | ----------- | | Basic Standing Task | 'h1' | | Basic Walking Task | 'jvrc_walk' | | Stepping Task (using footsteps) | 'jvrc_step' | | Cartpole swing-up | 'cartpole' | #### **To train:** ``` $ uv run run_experiment.py train --logdir <path_to_exp_dir> --num_procs <num_of_cpu_procs> --env <name_of_environment> ``` Note: Setting `RAY_ADDRESS=` ensures Ray starts a new local cluster instead of connecting to an existing one. #### **To play:** ``` $ uv run run_experiment.py eval --logdir <path_to_actor_pt> ``` Or, we could write a rollout script specific to each environment. #### **Cartpole** A minimal swing-up task for testing the RL pipeline. The goal is to swing the pole from hanging down to balancing upright. ``` $ uv run run_experiment.py train --env cartpole --n-itr 500 --std-dev 0.15 --learn-std --entropy-coeff 0.01 --minibatch-size 256 --max-traj-len 500 --no-mirror ``` ## Configu
Excerpt of 7,484 characters
Read on GitHub
141
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:6c4b028e43706cae, topic:reinforcement-learning, desc:reinforcement learning, readme:reinforcement learning