ForceFlow: Contact-Driven Robot Manipulation Data

ForceFlow is a dataset collected to support a force-aware reactive robot learning framework. It was created by author JokerESC and last updated on Hugging Face in May 2026. The data likely contains high-frequency contact dynamics for manipulation tasks.

Use Cases

• Training reactive robot policies based on contact-driven flow matching.
• Benchmarking force-aware manipulation algorithms for tasks like plug insertion.
• Studying high-frequency contact dynamics for actions like stamping or wiping.

Strengths

• Dataset is associated with a published research framework named ForceFlow.
• Focuses on a specific, challenging problem domain: contact-rich robot manipulation.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and dataset size are unknown, which may limit suitability assessment.

Provenance

Source

JokerESC on Hugging Face

Collection Method

Collected to support the ForceFlow research project.

Freshness

Last updated 2026-05-11 03:05:51; freshness should be verified.