Xiayu Zhao

Xiayu Zhao

Graduate student at University of Illinois Urbana-Champaign

Hello! 你好！

Welcome to my academic website! I’m Xiayu Zhao, a Master’s student in Computer Science at the University of Illinois Urbana-Champaign, with a PhD in Robotics and AI for Construction (January 2024 - present). I work in the Robotic, Automation, and Intelligent Sensing (RAISe) Lab under the supervision of Dr. Houtan Jebelli. I work closely with Dr. Yizhi Liu.

About My Research

My research focuses on AI and robotics for construction, with emphasis on:

  • Multimodal perception for robotic inspection: Computer vision and deep learning for automated inspection—including real-time roof defect segmentation (RRD-SegNet)—and sensor fusion for aerial–ground multi-robot systems.
  • AI and machine learning for human–robot collaboration: Grounding large language models in robot control for natural-language interaction, and LLM-driven safe navigation with terrain-adaptive path planning.
  • Construction robotics and cyber–physical systems: Hybrid UAV–UGV systems with interchangeable ground modules, adaptive kinematic and dynamic control for hexapods in complex environments, and reinforcement learning for robot navigation and UAV stability with payloads.
  • Worker safety and physiological monitoring: Transfer learning for heat-strain prediction and understanding exoskeleton effects on metabolic cost in repetitive construction tasks.
  • Digital fabrication and computational design: Computational design, algorithm optimization, and robotic fabrication (e.g., 3D concrete printing); earlier work on biomimetic and morphological simulation for design applications.

My work combines robotics, computer vision, machine learning, and human–computer interaction to improve construction safety, efficiency, and quality. I develop software interfaces for robotic construction, control and perception for UAV/UGV/hexapod systems, and methods that connect language models to robot control in construction environments.

Education

  • Doctor of Philosophy (PhD), Robotics and AI for Construction, University of Illinois Urbana-Champaign (January 2024 - present)
    • GPA: 3.96/4.00
    • Research: Robotics in Construction; AI for Robotics; Hybrid UAV-UGV Systems for Construction Tasks
    • Robotic, Automation, and Intelligent Sensing (RAISe) Lab
  • Master’s degree, Computer Science, University of Illinois Urbana-Champaign (August 2025 – May 2027)
    • GPA: 4.00/4.00
    • Focus: Mobile Robotics; Large Language Models; Deep Learning; Reinforcement Learning; Advanced Information Retrieval; Software Engineering
  • Master’s degree, Robotic Fabrication & 3D Concrete Printing， Tsinghua University, Beijing, China (September 2019 – July 2021)
    • GPA: 3.97/4.00
    • Research: Digital Fabrication, Computational Design, Algorithm Optimization, Machine Learning, Data Visualization, Game Development
  • Bachelor’s degree, Digital Fabrication & Parametric Design, Tianjin University, Tianjin, China (September 2014 – June 2019)
    • GPA: 3.86/4.00
    • Focus: Computational Design for Architecture

Research Experience

I am currently a PhD student at the University of Illinois Urbana-Champaign, working on several projects in the RAISe Lab as part of my research focus in endowing construction robots with multi-modal perception and native language reasoning including:

  • Real-time roof defect segmentation using deep learning (RRD-SegNet)
  • Adaptive intelligence for robot navigation efficiency using deep reinforcement learning
  • Grounding large language models in robot control for human-robot collaboration

Previously, I worked as a Research Engineer at the Institute of Future Human Habitats at Tsinghua University, where I engaged large-scale robotic construction technologies and software interfaces for robotic control, with research in digital fabrication, computational design, and algorithm optimization.

News

  • June 2025: Tianyu Ren, Xiayu Zhao, and Dr. Houtan Jebelli were recognized as one of the six Olympiad Medalists across all engineering fields at the 2025 Olympiad in Engineering Science, held at the University of Stavanger in Norway, for their submission titled “Autonomous Drone-based System for Precision, Non-contact Surface Finishing in Construction.”
  • June 2025: Xiayu Zhao presented paper “AI-enhanced Adaptive Landing Gear for Aerial Robots in Dynamic Environments,” at the 2025 Olympiad in Engineering Science (OES), Stavanger, Norway.
  • May 2025: Xiayu Zhao was named the First Place Winner of the 2025 ASCE Global Robotics and Automation Competition for the work in construction robotics at i3CE 2025. Congratulations!
  • March 2025: Journal paper by Xiayu Zhao and Houtan Jebelli on “A computational method for real‐time roof defect segmentation in robotic inspection,” was published has been featured as the cover image in Computer‐Aided Civil and Infrastructure Engineering.
  • Multiple papers accepted to the 2025 International Conference on Computing in Civil Engineering (i3CE)
  • Multiple papers accepted to the 2024 International Conference on Computing in Civil Engineering (i3CE)