I am an incoming Ph.D. student at Northeastern University (Foshan), under the supervision of Prof. Xiaoguang Ma. Recently, I am focusing on research in Agentic UAVs. Prior to this, I received my Master’s degree in Electronic Information from Zhejiang University of Technology in June 2024, where my work was dedicated to developing visual Simultaneous Localization and Mapping (vSLAM) algorithms for ground mobile robots to achieve visual navigation in dynamic scenes.

More about me: Email / Google Scholar / Github / Curriculum Vitae

Feel free to contact me by email if you are interested in discussing or collaborating with me.

📚 Research Vision

My current research focuses on developing self-improving / self-evolving embodied navigation agents, and exploring their applications in aerial and ground robots. In the long run, my research vision is to build an embodied agent that can autonomously perceive, make decisions, and learn continuously like humans. To achieve this, I primarily focus on:

• Language-driven visual navigation (e.g., VLN and ObjectNav).
• Large foundation models (e.g., LLMs and VLMs) and their training techniques.
• Lifelong learning of LLM / VLM-powered agents.
• Vision-Language-Action models and their applications.
• World models and their applications.

🔥 News

📝 Selected Publications

A Survey of Self-Evolving Embodied Agents: On the Path to Real-World Physical Intelligence

Xiangyu Dong, Zirong Zhang, Maozhi Li, Huanhuan Ke, Xiaoguang Ma ^†.

Introduction

• This survey examines self-evolving embodied agents as a route to real-world physical intelligence. It defines self-evolution through persistent growth from interaction, memory, feedback, reflection, and updating. It proposes a closed-loop framework, reviews technical progress and mechanisms, and discusses benchmarks, evaluation protocols, safety governance, challenges, and directions for open-world embodied AI.

• [IROS 2026] ViSA-Enhanced Aerial VLN: A Visual-Spatial Reasoning Enhanced Framework for Aerial Vision-Language Navigation, Haoyu Tong *, Xiangyu Dong *, Haoran Zhao, Yaoming Zhou ^†, Xiaoguang Ma ^†.
• SE-VLN: A Self-Evolving Vision-Language Navigation Framework Based on Multimodal Large Language Models, Xiangyu Dong *, Haoran Zhao *, Jiang Gao, Haozhou Li, Xiaoguang Ma ^†, Yaoming Zhou ^†, Fuhai Chen, Juan Liu.
• SM-VLN: A Self-Evolving Multi-Agent Vision-Language Navigation Framework Based on Foundation Models, Xiangyu Dong, Haozhou Li, Xiaoguang Ma ^†, Wei Meng, Zhengtao Hu.
• CMMR-VLN: Vision-and-Language Navigation via Continual Multimodal Memory Retrieval, Haozhou Li *, Xiangyu Dong *, Huiyan Jiang, Yaoming Zhou, Xiaoguang Ma ^†.
• PM-Nav: Prior-Map Guided Embodied Navigation in Functional Buildings, Jiang Gao *, Xiangyu Dong *, Haozhou Li, Haoran Zhao, Yaoming Zhou, Xiaoguang Ma ^†.
• LightZeroNav: Zero-Shot Vision Language Navigation in Continuous Environments Based on Lightweight VLMs, Kun Luo *, Xiangyu Dong *, Xiaoguang Ma ^†., Haoran Zhao, and Yaoming Zhou ^†.
• CLOSER-VLN: Closed-Loop Self-Verified Retrieval-Augmented Reasoning for Aerial Vision-Language Navigation, Shaoxuan Li *, Xiangyu Dong *, Xiaoguang Ma ^†, Junfeng Chen, Haoran Zhao, Yaoming Zhou ^†.

(co-first authors: *, corresponding authors: †)

🎖 Honors and Awards

• 2023: Second-Class Academic Scholarship, Zhejiang University of Technology.
• 2022: First-Class Academic Scholarship, Zhejiang University of Technology.
• 2022: Second Prize in the East China Division, China Graduate Electronic Design Competition.
• 2021: Freshman Scholarship, Zhejiang University of Technology.
• 2021: Excellent Undergraduate Graduation Thesis, Nantong University.
• 2021: Outstanding Student, School of Transportation and Civil Engineering, Nantong University.
• 2019: First Prize in the East Division, National College Student Embedded Chip and System Design Competition.
• 2019: Second Prize, Jiangsu Provincial “Internet +” College Student Innovation and Entrepreneurship Competition.
• 2019: Second Prize, Nantong University “Internet +” College Student Innovation and Entrepreneurship Competition.

📖 Educations

• 2026.09 - Expected: Ph.D., College of Information Science and Engineering, Northeastern University, Shenyang, China.
• 2021.09 - 2024.06: Master, College of Information Engineering, Zhejiang University of Technology, Hangzhou, China.
• 2017.09 - 2021.06: Bachelor, School of Transportation and Civil Engineering, Nantong University, Nantong, China.

💬 Reports

• 2023.05: “A Robust Visual SLAM Based on Key Point Instantaneous Rate Identification in Dynamic Environments”, 2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS).

💻 Internships

• 2025.05 - Present: Beihang University–Tianmushan Laboratory, co-supervisor: Professor Yaoming Zhou.
• 2024.05 - 2026.08: Northeastern University, supervisor: Professor Xiaoguang Ma.

🛠 Social Services

• Reviewer: AAAI 2026, IROS2026.