sislaboratory

owm-iss-noncoop-goal-dt50ms

Docking approaches to the International Space Station with a Dragon-class chaser spacecraft maneuvering from 100-500m under rigid-body dynamics with non-cooperative target navigation error. Generated for world-model training at 20 Hz with 502,771 training transitions and 52,373 validation transitions.

Downloads0
Episodes113
Hours7.71

Why This Matters for Physical AI

This dataset demonstrates world-model training on high-dimensional spacecraft dynamics with realistic sensor degradation and non-cooperative target constraints, advancing embodied AI for autonomous rendezvous and docking in space operations.

Technical Profile

Modalities
rgblidarproprioceptionforce_torque
Robot Embodiments
spacecraftDragon-class chaser
Action Space
force_torque
Environment
simulationspace
Task Types
dockingspacecraft_rendezvousnavigation
Episodes
113
Total Hours
7.71
Data Format
LeRobot
Annotation Types
language_instructionsreward_labels
Part of the Out of This World Model family

Access

Need custom rgb data?

Claru builds purpose-built datasets for simulation applications with dense human annotations and quality assurance.

Request a Sample Pack

Related Datasets