Datasets:

ReForceMind
/

quadruped_domain_randomization

language:
  - en
license: mit
tags:
  - robotics
  - test-time-adaptation
  - reinforcement-learning
  - continuous-control
  - flow-matching
dataset_info:
  features:
    - name: seed
      dtype: int32
    - name: step
      dtype: int32
    - name: joint_torques
      sequence: float32

Dataset Card for quadruped_domain_randomization

Dataset Description

This dataset contains Proportional-Derivative (PD) corrective joint torques generated during Test-Time Adaptation (TTA) simulations for the Quadruped environment using the domain_randomization policy. The dataset is structured for use in downstream machine learning workflows (such as training secondary diffusion or flow-matching models).

Physical Environment

Robot: Quadruped
Degrees of Freedom (DOF): 12

Normalization

The continuous joint_torques have been standardized globally to zero-mean and unit-variance across all seeds. To un-normalize the data back to raw physical torque values, use the following constants:

Mean ($\mu$): [0.5391409993171692, -0.9164595007896423, 4.03103494644165, -0.5374435782432556, -0.9164064526557922, 4.03103494644165, 0.003860426601022482, 2.696013927459717, 4.032361030578613, 0.08484119921922684, 2.695657253265381, 4.032401084899902]
Standard Deviation ($\sigma$): [0.06029755249619484, 0.12463501840829849, 0.16779327392578125, 0.05818825960159302, 0.12457720190286636, 0.16778984665870667, 0.018216369673609734, 0.16376766562461853, 0.17438504099845886, 0.030601857230067253, 0.16376431286334991, 0.17441211640834808]

Usage

from datasets import load_dataset
dataset = load_dataset("elprofesoriqo/quadruped_domain_randomization")