R2HandoverSim: A Simulation Framework and Benchmark for Robot-to-Human Object Handovers

2026-06-18

AI Summary

The absence of standardized evaluation protocols has hindered objective comparisons of robotic methods for handing over objects to humans. This work addresses this gap by introducing a simulation benchmark specifically designed for robot-to-human (R2H) handover interactions. It proposes five complementary metrics—planning feasibility, reachability, grasp stability, manipulability, and safety—to comprehensively quantify handover quality, and systematically evaluates four baseline methods for shared grasp pose prediction. A user study involving 30 participants demonstrates a strong correlation between simulation results and real-world performance, validating that the proposed benchmark enables reproducible, objective, and perceptually consistent evaluation of R2H interactions.

Abstract

We present R2HandoverSim, a simulation benchmark for robot-to-human (R2H) object handovers. Although R2H handover methods have advanced rapidly, the lack of standardized evaluation protocols impedes objective comparison. Our benchmark enables reproducible evaluation by systematically comparing four baselines on their predicted shared grasp poses. We conduct a user study with 30 participants, analyze baseline performance, and show that simulation results correlate with real-world evaluation outcomes. Crucially, five complementary metrics (planning feasibility, reachability, grasp stability, grasp affordance, and safety) better reflect user-perceived handover quality than overall success rate alone. Website and code: https://robot-future.github.io/r2handoversim/.

Problem

• robot-to-human handover
• simulation benchmark
• standardized evaluation
• grasp pose
• handover quality

Innovation

• robot-to-human handover
• simulation benchmark
• shared grasp pose
• evaluation metrics
• user study