Downloads · 30 days
4
11% of all-time downloads
han-humanoid-navigation-v1
Caplin43/han-humanoid-navigation-v1
han-humanoid-navigation-v1 is a reinforcement learning model from Caplin43. Use it for the reinforcement learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.
This model is designed for humanoid robot balance control using Proximal Policy Optimization (PPO). It predicts joint torque adjustments to maintain stability during standing and walking.
All-time downloads
36
Public
Repo size
—
Likes
0
Public
Repo makeup
Click a slice to open those files.
Other1.5 KB · 54%
At a glance
- Task
- Reinforcement Learning
- License
- mit
- Model type
- ppo
- Access
- Public
- Created
- Feb 14, 2026
- Updated
- Feb 14, 2026
- SHA
- 1acaa5d7
Model card
From the Hugging Face model README
HAN Humanoid Balance PPO v1
Overview
This model is designed for humanoid robot balance control using Proximal Policy Optimization (PPO). It predicts joint torque adjustments to maintain stability during standing and walking.
Architecture
- Algorithm: Proximal Policy Optimization (PPO)
- Actor-Critic Network
- 3 Hidden Layers (256 units each)
- Activation: ReLU
Training Environment
- Physics-based humanoid simulation
- Variable terrain conditions
- Randomized push disturbances
Observation Space
- Joint positions
- Joint velocities
- IMU orientation
- Ground contact signals
Action Space
- Continuous joint torque outputs
Intended Use
- Humanoid robot balance research
- Reinforcement learning experiments
- Robotics simulation projects
Limitations
This model is trained in simulation and may require fine-tuning for real-world deployment.
Author
Caplin43
- Task
- Reinforcement Learning
- Type
- ppo
- License
- mit
- Created
- Feb 14, 2026
- Updated
- Feb 14, 2026