Collaborative Object Handover in a Robot Crafting Assistant

2025-02-27

AI Summary

This work addresses safe and natural object handover in human–robot collaboration (HRC) during manual crafting tasks. We present the first end-to-end, context-aware handover policy explicitly modeling human behavioral cues—including action intent, body pose, and interaction rhythm—within real-world crafting scenarios. The policy is trained via behavior cloning from teleoperation data and supports dynamic intent adaptation and rhythmic synchronization with the human partner. Evaluated through cross-validation and a user study, our approach significantly outperforms baseline methods in handover success rate, safety, and perceived naturalness; users rated interactions as comparable to human–human collaboration. Furthermore, our analysis identifies critical optimization avenues: precise handover timing estimation and fine-grained pose coordination. This work establishes a scalable, empirically grounded modeling framework for context-aware handover in HRC, advancing both theoretical understanding and practical deployment of collaborative manipulation systems.

Technology Category

• Intelligent Robots: Manipulation
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: The sharing economy
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Robots are increasingly working alongside people, delivering food to patrons in restaurants or helping workers on assembly lines. These scenarios often involve object handovers between the person and the robot. To achieve safe and efficient human-robot collaboration (HRC), it is important to incorporate human context in a robot's handover strategies. Therefore, in this work, we develop a collaborative handover model trained on human teleoperation data collected in a naturalistic crafting task. To evaluate the performance of this model, we conduct cross-validation experiments on the training dataset as well as a user study in the same HRC crafting task. The handover episodes and user perceptions of the autonomous handover policy were compared with those of the human teleoperated handovers. While the cross-validation experiment and user study indicate that the autonomous policy successfully achieved collaborative handovers, the comparison with human teleoperation revealed avenues for further improvements.

Problem

• Develop collaborative handover model
• Incorporate human context in HRC
• Evaluate autonomous handover policy performance

Innovation

• Human-robot collaborative handover model
• Trained on human teleoperation data
• Cross-validation and user study evaluated