Biography

I’m an MS student at CMU RI, working at LeCAR Lab advised by Prof. Guanya Shi.

Previously, I received my B.Eng. from Shanghai Jiao Tong University (SJTU), where I was advised by Prof. Hesheng Wang and Prof. Danping Zou. After that, I worked with Prof. Richard Ren remotely during a summer research. Subsequently, I worked with Prof. Huazhe Xu as a research assistant at Tsinghua University (IIS).

During my undergraduate career, I led the programming group at SJTU VEX Robotics Club. Additionally, I have participated other types of competitions related to Unmanned Vehicles (UV) and Unmanned Aerial Vehicle (UAV).

I’m interested in Robotics, Machine Learning and Optimal Control. Currently, I’m working on Humanoid and Aerial Manipulation.

Education

Carnegie Mellon University Aug. 2024 - Present
Master Student
Major: Robotic Systems Development
Shanghai Jiao Tong University Sep. 2019 - Jun. 2023
Undergraduate Student
Major: Automation

Experience

Full-time Research Assistant Jan. 2024 - Jul. 2024
Tsinghua Embodied AI Lab (TEA Lab)
Advisor: Prof. Huazhe Xu
Topic: Mobile Manipulation

Research Intern (Remote) Summer 2023
Robotics Autonomy and Planning Lab at SJTU
Advisor: Prof. Zhongqiang Ren
Topic: Multi-Agent Path Finding

Undergraduate Research Assistant Jan. 2023 - Jun. 2023
Intelligent Robotics and Machine Vision Lab at SJTU
Advisor: Prof. Hesheng Wang
Topic: Image-Based Visual Servo for Quadrotor

Research

  • Multi-Agent Combinatorial Path Finding with Heterogeneous Task Duration
    Yuanhang Zhang, Xuemian Wu, Hesheng Wang, Zhongqiang Ren

    TL;DR
    Proposed two conflict-based search methods—CBSS-TPG (conflict-free) and CBSS-D (conflict-free and minimal cost) to solve the multi-agent combinatorial path finding problem with target duration.

    Image 0

  • ViTaS: Visual Tactile Soft Fusion Contrastive Learning for Reinforcement Learning
    Yufeng Tian†, Shuiqi Cheng†, Tianming Wei, Tianxing Zhou, Yuanhang Zhang, Zixian Liu,
    Zhecheng Yuan, Huazhe Xu

    • Under Review, 2024

    TL;DR
    Introduced ViTaS, a framework using using Soft Fusion Contrastive Learning and a CVAE module to fuse visual and tactile information, achieving state-of-the-art performance with high sample efficiency and low parameter count in robotic manipulation.

    Image 0

  • Perception-constrained Visual Servoing Based NMPC for Quadrotor Flight

    TL;DR
    Incorporated quadrotor dynamics and visual feature dynamics into NMPC to enable the quadrotor to flight purely based on visual information without localization.

    Image 0


Competitions

UAV Intelligent Perception Technology Competition
National Third Prize (10%)
Shanghai, China
2022.09-2022.11
  • As the team leader, I implemented a SE(3) controller for quadrotor control in PX4-Autopilot, deployed YOLOv5 with TensorRT for object detection and P3P for pose estimation, and used RAPIDDS to generate collision-free trajectories for quadrotor navigation.
Image 0 Image 1
Image 0 Image 1
National University IOT Design Competition
National First Prize & Harmony Innovation Award (1%)
Shanghai, China
2022.06-2022.09
  • As the team leader, I led our team of 4 to design ‘HarClass’ from scratch, an app for modern intelligent classrooms, utilizing the distributed features of HarmonyOS, leveraged BearPi-HM Micro for environment awareness, and developed a data visualization website using custom communication protocols for cloud connectivity. [Video]
Image 2 Image 3
National University ICT Competition (Innovation)
National Second Prize (5%)
Shanghai, China
2021.09-2022.01
  • As the team leader, I conducted data set collection from edge devices via Socket and utilized the MindSpore framework along with the Ascend chip for the model combination of target detection and controlling. Finally, I deployed the combined model in an automated vehicle. [Video]
Image 2 Image 3


Awards

SJTU Outstanding Graduate (top 5%)

SJTU Merit Student (top 3%)

SJTU Academic Progress Scholarship

SJTU Merit Scholarship

Leadership

SJTU-VEX
Programming Team Leader
Shanghai, China
2020.03-2023.01
  • Managed a team of 20+ undergraduates to develop algorithms for custom vehicle applications, achieving one of the highest program level among Chinese universities as well as winning lots of national/international champions.
Image 0 Image 1 Image 2
  • Led the development of SJTU VEX’s AI automation system, including global localization, visual recognition, and communication modules and presented results to universities and IFI Chinese representatives.
Image 0 Image 1


Service

  • Reviewer: ICRA 2025