Haruka Matsuo
Papers
2
Total Citations
7
H-Index
2
About
Haruka Matsuo is a rising researcher in domestic service robotics and human-safe robot interaction, whose work centers on enabling robots to perceive, predict, and communicate risks during object manipulation and handover tasks. Her most-cited paper introduces a shared Transformer encoder with a mask-based 3D model estimation framework for container mass estimation—a critical contribution for safe human-to-robot handovers, where accurately inferring the physical properties of containers and their fillings prevents accidents. Building on this, her 2024 work pioneers "nearest neighbor future captioning," a novel approach that generates natural language descriptions of potential collisions before they occur during object placement tasks. This advances linguistic explainability in domestic service robots, allowing them to proactively describe future risks from their own actions—a capability previously underexplored. With her papers accumulating early citations, Matsuo is establishing herself as a key voice in robot transparency and proactive safety, bridging perception, prediction, and human-understandable communication to make autonomous robots more trustworthy in everyday environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2