PhysDBPhysical AI Map

Data

Robot data

Robot trajectories, observations, actions, labels, task descriptions, and environment metadata used to train or evaluate policies.

datasettrajectory

What it is

Robot data differs from web text or images because it must encode action and physical interaction.

Why it matters

Data scarcity and heterogeneity are central constraints for general robot learning.

How not to overread it

More data is not automatically better if embodiments, tasks, sensors, or labels are incompatible.

Related edges

expanded by

Cross-embodiment data

Dataset composition

Cross-embodiment is not automatic cross-hardware reliability.

organizes

LeRobot

Open robot learning workflows

Tooling is not a benchmark result.

augments

Synthetic data

Dataset expansion

Synthetic data must not be counted as real-world proof.