EMBODIED INTELLIGENCE

Ronghe QIU

Ph.D. Candidate · HKUST (Guangzhou)

I am a Ph.D. candidate in Artificial Intelligence at the Hong Kong University of Science and Technology (Guangzhou), advised by Junwei Liang and co-supervised by Ying-Cong Chen. I obtained my M.Phil. degree in Artificial Intelligence from the same university in 2024 under the supervision of Junwei Liang. My research focuses on embodied AI for long-horizon mobile manipulation. I am particularly interested in how robots can reason through complex tasks and generalize to new tasks and environments.

I received my bachelor's degree in Finance from Sun Yat-sen University and worked in finance for four years, primarily advising companies on IPOs in the U.S. market.

Inspired by the belief that technology can change the world for the better, I transitioned to AI research. I hope to help build a future where more people can live in peace and happiness.

  • Embodied agents
  • Long-horizon mobile manipulation
  • Human-centered robotics
Portrait of Ronghe QIU
Beijing & Guangzhou, China

01News

02Publications

2026
CoRL 2026Accepted

Look Before You Move: Agent-Interface-Aligned Supervision for Open-World Mobile Manipulation

Ronghe Qiu, Teli Ma, Jiaming Zhou, Zeying Gong, Yiting He, Qiang Huang, Junwei Liang

03Projects

Agentic Robot Tour Guide

  • Developed automated mapping and relocalization using point cloud features for navigation in dynamic indoor environments.
  • Designed a skill-based agent framework in which a VLM calls navigation, speech, and gesture actions in response to user commands.
  • Enabled multimodal interaction through real-time text-to-speech and internet-connected knowledge retrieval.

TTS / Scene graphs / VLM reasoning

Multi-Floor Robot Navigation

  • Deployed the VLFM algorithm on Unitree Go2, bridging the simulation-to-reality gap for mapping and path planning.
  • Integrated navigation and elevator manipulation into a unified pipeline.

SLAM / Object detection / VLM reasoning