HARP-Bench: Benchmarking Robot Manipulation Around Active Humans
HARP-Bench: Benchmarking Robot Manipulation Around Active Humans
This research introduces HARP-Bench, a benchmark designed to evaluate robot manipulation policies in workspaces shared with active humans.
The paper introduces HARP-Bench, a new benchmark for evaluating robot manipulation policies in workspaces shared with active humans, organized into four conditions (Cueing, Contribution, Interruption, Co-presence) and supported by simulation training demonstrations and initial real-world validation on one task family.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- code: Available Artifact link Open source S6
What remains unknown
Dataset licensing and commercial-use rights are not established here.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
A real-world study on one task family shows improved performance under human contribution, but not interruption, providing initial evidence of the data's utility and identifying limits of transfer. Code is available at https://github.com/UMass-Embodied-AGI/harp.
Open source S6
Implications for data suppliers
RoboSignal interpretation and collection questions, not statements of buyer demand.
- Confirm the required data type and collection setting with the buyer; this source does not establish a complete collection specification.
- Validate demand and acceptance criteria with a buyer before scaling. Publication, popularity and a research result do not establish a purchase commitment.
Source：arXiv Robotics — research abstracts · arxiv.org