Aodi Wu

Aodi Wu

Building reliable AI systems for on-orbit servicing and space robotics.

About

I am a Ph.D. student in Computer Application Technology at the University of Chinese Academy of Sciences, working at the Technology and Engineering Center for Space Utilization, Chinese Academy of Sciences.

My research focuses on space AI, especially for on-orbit servicing, multimodal spacecraft perception, and embodied intelligence. I am interested in building systems that connect perception, reasoning, control, simulation, and real-world validation.

More broadly, I care about research that moves from isolated algorithms to complete engineering systems. My recent work includes MCP-based space agents, large-scale multimodal benchmarks for spacecraft perception, and cross-domain methods for space imagery understanding.

News

  • 2026.03 Refreshing this website to better present my work on space AI, embodied agents, and multimodal spacecraft perception.
  • 2025.10 Won 2nd place in the IROS 2025 RoboSense Challenge.
  • 2025.07 Published SpaceMind, an MCP-based agent architecture fusing large and small models for on-orbit servicing, at IAA-SPAICE 2025.
  • 2025 Published a paper on cross-domain spacecraft component segmentation at ICDIP 2025.
  • 2024.03 Won 1st place in the CVPR 2024 SPARK Spacecraft Pose Estimation Challenge.
  • 2024.03 Ranked 4th in the CVPR 2024 SPARK Spacecraft Part Segmentation Challenge.
  • 2023 Received an authorized patent on intelligent in-orbit exposure and focus control for space cameras.

Selected Publications

  • SpaceMind: An MCP-based Agent Architecture Fusing Large and Small Models for On-orbit Servicing. IAA-SPAICE 2025.
    [Code]

  • Cross-domain Spacecraft Component Segmentation Based on Edge Consistency Generative Neural Network. ICDIP 2025.
    [Code]

  • Non-cooperative Spacecraft Relative Navigation Based on Monocular Camera in Space On-orbit Servicing. ICoSR 2022.

Patent

  • Authorized Patent (2023): Intelligent in-orbit exposure and focus control method, system, and electronic equipment for space cameras.

Honors and Awards

  • 2025.11 1st Place, UCAS Graduate Forum, Aerospace Session.
  • 2025.10 2nd Place, IROS 2025 RoboSense Challenge.
  • 2024.03 1st Place, CVPR 2024 SPARK Spacecraft Pose Estimation Challenge.
  • 2024.03 4th Place, CVPR 2024 SPARK Spacecraft Part Segmentation Challenge.
  • 2016-2020 National Encouragement Scholarship and recommendation for graduate study at Nanjing University of Aeronautics and Astronautics.

Selected Projects

  • SpaceMind: An MCP-based space agent architecture for perception, decision-making, and control in on-orbit servicing.
    [GitHub]

    SpaceMind method figure

  • SpaceSense-Bench / MM-SPACE: A large-scale multimodal benchmark for spacecraft perception and pose estimation, built on UE5 and AirSim.
    [Repository] [Paper] [Project Page] [Dataset]

    SpaceSense-Bench teaser figure

  • Cross-domain Spacecraft Segmentation: Domain-gap-aware segmentation for spacecraft parts with cross-domain generalization.
    [GitHub]

  • In-orbit Camera Intelligence: Engineering work on target detection, relative navigation, and intelligent exposure/focus control for real space cameras and on-orbit validation.

Education

  • 2023.09 - Present — Ph.D. in Computer Application Technology, University of Chinese Academy of Sciences.
  • 2020.09 - 2023.06 — M.S. in Computer Application Technology, University of Chinese Academy of Sciences.
  • 2016.09 - 2020.06 — B.E. in Detection Guidance and Control Technology, Nanjing University of Aeronautics and Astronautics.

Research Interests

  • Space AI and on-orbit servicing
  • Multimodal spacecraft perception
  • Tool-use agents and embodied intelligence
  • Relative navigation and autonomous control
  • Simulation-to-real system validation

Contact