timestamp float64 0.01 0.05 | joint_angle_1 float64 11.8 14 | joint_angle_2 float64 7.9 9 | imu_pitch float64 -0.04 0.08 | imu_roll float64 -0.05 0.02 | foot_pressure_left float64 44 47.2 | foot_pressure_right float64 42.8 46.5 | balance_action_label stringclasses 3
values |
|---|---|---|---|---|---|---|---|
0.01 | 12.5 | 8.2 | 0.02 | -0.01 | 45.2 | 44.8 | stable |
0.02 | 13.1 | 8.5 | 0.05 | -0.03 | 46.1 | 43.9 | adjust_left |
0.03 | 11.8 | 7.9 | -0.04 | 0.02 | 44 | 46.5 | adjust_right |
0.04 | 12.9 | 8.1 | 0.01 | -0.02 | 45.5 | 45 | stable |
0.05 | 14 | 9 | 0.08 | -0.05 | 47.2 | 42.8 | adjust_left |
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
HAN Humanoid Balance Control Dataset
Overview
This dataset is designed for humanoid robot balance and posture stabilization training.
Description
The dataset contains joint sensor readings and balance correction labels collected from humanoid robot simulations.
Data Structure
- timestamp
- joint_angle_1
- joint_angle_2
- imu_pitch
- imu_roll
- foot_pressure_left
- foot_pressure_right
- balance_action_label
Task Type
Supervised Learning – Classification
Use Case
- Balance correction training
- Fall prevention modeling
- Posture stability improvement
Size
10,000 samples (simulated data)
License
MIT
- Downloads last month
- 16