Human-to-Robot Handovers

Overview

Human-to-Robot Handovers of Unknown Objects is a track of the 12th Robotic Grasping and Manipulation Competition (RGMC) at the 2027 IEEE/RAS International Conference on Robotics and Automation (ICRA) in Seoul, South Korea.

Key features

• Dynamic handovers of previously unseen objects
• Real-time multimodal perception of object properties during the handover
• An affordable setup without motion capture, markers, or prior knowledge of specific object models
• Remote qualification using the CORSMAL Benchmark protocol

Track description

The real-time estimation through perception of the physical properties of objects manipulated by humans is important to inform the control of robots and reason and perform accurate and safe grasps of objects handed over by humans. However, estimating the physical properties of previously unseen objects using only an inexpensive and affordable setup is challenging due to illumination variations, transparencies, reflective surfaces, and occlusions caused both by the human and the robot.

Goal: Assessing the generalisation capabilities of robots when handing over previously unseen objects (e.g. food containers or drinking glasses filled or not with unknown content, or other objects). No object properties are initially known to the robot (or the team) that must infer these properties on-the-fly through perception of the scene and adapt their actions during dynamic human-to-robot handovers.

Opportunities

Beyond the chance of winning the track, participants have the opportunity to:

• Test algorithms on their own robot or on robots available at the competition site
• Earn recognition at ICRA 2027
• Share your solution and findings with the community, including through a research article
• Showcase a demo video on the RGMC YouTube channel
• Receive feedback from the competition organisers and peers
• Promote open science and reproducibility by releasing code and datasets
• Network and collaborate with researchers and practitioners

Further opportunities may be announced as the programme develops.

Qualification phase

Teams will undertake a remote qualification phase in their laboratories. Further details on qualification requirements and the timeline will be announced.

See the qualification page for more information.