Jiarui Li

I'm a senior undergrad at the College of Engineering at Peking University, majoring in robotics engineering. I am grateful to be advised by Professor Heng Yang and to be working as a visiting undergraduate researcher at the computational robotics research group at the School of Engineering and Applied Sciences (SEAS), Harvard University.

I'm applying for graduate programs now! I will receive my B.E. from Peking University in 2024. After that, I would like to continue my journey in the beautiful world of robotics. 🤖️🐱

 |  Email  |  Google Scholar  |  LinkedIn  |  CV  |  X (Twitter)  | 

profile photo


News

Dec 1, 2023 Extended version of the IROS Best Paper Finalist on Mobile Manipulation Paper on aerial manipulation, titled Design, Planning, and Control of an Over-actuated Aerial Manipulator for Sequential Manipulation, submitted to Transactions on Robotics (T-RO)
Sep 28, 2023 Thrilled to announce that my paper Sequential Manipulation Planning for Over-actuated Unmanned Aerial Manipulators on aerial manipulation at IROS 2023 won the Best Paper Finalist Award on Mobile Manipulation! Looking forward to presenting it at IROS!
Sep 20, 2023 Happy to announce that I have received the National Scholarship (top 0.2% among all the undergraduate students) and been honored with the title of Pacemaker to University Merit Student by Peking University for 2023.
Jul 3, 2023 Hello Boston! Grateful to be advised by Professor Heng Yang to start working as a visiting undergraduate researcher at the computational robotics research group at the SEAS, Harvard University.
Jun 22, 2023 Two papers Sequential Manipulation Planning for Over-actuated Unmanned Aerial Manipulators and Aggregating Single-wheeled Mobile Robots for Omnidirectional Movements accepted to 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2023)
Jun 16, 2023 One paper L³F-TOUCH: A Wireless GelSight with Decoupled Tactile and Three-axis Force Sensing accepted to IEEE Robotics and Automation Letters (RA-L)

Research Overview

My research focuses on creating trustworthy robot systems that that can move in the real world and interact with humans and environments stably, safely, and robustly.
researchstatement
My past research spans a variety of Real Robotic Systems, including aerial robots, ground vehicles, and manipulators. I accumulated rich hands-on experience in developing real robot system in both hardware and software. These experiences have granted me invaluable insights highlighting the safety, stability, and robustness issues of real robots. Such challenges motivated me to delve into the theoretical foundations of robot systems, aiming to develop trustworthy autonomy with guaranteed, even optimal, performance. Now I am actively exploring Theory-Based Data-Driven Control, ranging from approximate dynamic programming to diffusion-based methods, using learning-based methods to seek reliable and effective control solutions for non-linear systems. In my future research, I hope to continue delving deeper into creating trustworthy robot systems leveraging data-driven methods, including reliable perception, planning and control, and apply these on real robots for applications and validations.

Publications

*indicates joint first authors, Representative papers are highlighted

iros2023aerialvkc Sequential Manipulation Planning for Over-actuated Unmanned Aerial Manipulators
Yao Su*, Jiarui Li*, Ziyuan Jiao*, Meng Wang, Chi Chu, Hang Li, Yixin Zhu, Hangxin Liu
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2023
Extended version submitted to Transactions on Robotics (T-RO)
Project Page / paper / video

Best Paper Award Finalist on Mobile Manipulation at IROS 2023!

Tackled sequential manipulation planning for UAMs, considering coordinated motions of the base, manipulator, and object. Using a VKC-based motion planning framework and a hierarchical control framework, we for the first time show new sequential manipulation capabilities for an over-actuated UAM in simulation and experiment.

iros23wmr Aggregating Single-wheeled Mobile Robots for Omnidirectional Movements
Meng Wang* Yao Su*, Hang Li, Jiarui Li, Jixiang Liang, Hangxin Liu
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2023
paper / video

Our novel modular robot system self-reconfigures and moves omnidirectionally through magnetic docking. The modules have steerable omni wheels and magnets on each corner. An optimization-based method distributes wheel heading direction for efficient movement. We validate the system's trajectory tracking and demonstrate stable caging formation and smooth transportation.

ral23gelsight L³F-TOUCH: A Wireless GelSight with Decoupled Tactile and Three-axis Force Sensing
Wanlin Li*, Meng Wang*, Jiarui Li, Yao Su, Devesh K. Jha, Xinyuan Qian, Kaspar Althoefer, Hangxin Liu
IEEE Robotics and Automation Letters (RA-L), 2023
paper / video

Designed and built the L³F-TOUCH sensor, an enhanced version of the classic GelSight sensor, to acquire a much better three-axis force sensing capability while being light-weight, low-cost and wireless for the ease of replication and deployment.

iros2022downwash Downwash-aware Control Allocation for Over-actuated UAV Platforms
Yao Su*, Chi Chu*, Meng Wang, Jiarui Li, Yang Liu, Yixin Zhu, Hangxin Liu
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2022
project page / paper / video

Leveraging the input redundancy in over-actuated UAVs, we tackle downwash effects between propellers with a novel control allocation framework that explores the entire allocation space for an optimal solution that reduces counteractions of airflows.


Experiences & Honors
pku_seal

Harvard University
Visiting Undergraduate Researcher in SEAS

Jul 2022 - Present, Boston, MA

Research Advisor: Prof. Heng Yang

pku_seal

Beijing Institute for General Artificial Intellignece (BIGAI)
State Key Laboratory for Cross-Media AGI
Undergraduate Research Intern in Robotics Lab

Jan 2022 - Jun 2023, Beijing, China

Research Advisor: Dr. Yao Su, Dr. Hangxin Liu
● Intern Outstanding Contribution Award from BIGAI in 2023

pku_seal

Peking University
Bachelor of Engineering in Robotics Engineering
Sep 2020 - Present, Beijing, China

● National Scholarship (highest scholarship for chinese college students) in 2023
● Pacemaker to University Merit Student in 2023
● University Merit Student in 2022
● Schneider Scholarship from College of Engineering in 2022


Selective Projects
aerialrobotics Aerial Robotics: UAV & UAM's Design, Control and Planning
Beijing Institute for General Artificial Intelligence (BIGAI), Robotics Lab
advisor: Dr. Yao Su, Dr. Hangxin Liu, Prof. Yixin Zhu, Prof. Song-Chun Zhu

Designed and built a fully-actuated UAV platform composed of 2-DoF passive joint with commercial quadcopter; Implemented an controller to avoid the downwash disturbance during flipping motion; Further designed and built a novel aerial manipulator, and implemented the planning and control algorithms to perform challenging indoor sequential task (such as installing a light bulb on the ceiling).

Accumulated experiences in mechatronic design and the implementation of control and planning algorithms. These hardware-related experiences can help me deal with problems like algorithm implementation efficiently.

scalefreenetwork Collective Intelligence: Evolutionary Game Theory, Game & Cooperation on Complex Networks
Peking University, College of Engineering
advisor: Prof. Aming Li, Dr. Lecheng Ruan, Prof. Long Wang

We use Monte Carlo Simulation to explore multi-agent intelligence and game strategy's evolution on complex networks, including the Scale-Free network and Erdős-Rényi network, to explain the emergence of cooperation in sizeable structured groups like human society, where the Prison Dilemma is quite common. Compare the effects of different parameters and try to explain this phenomenon clearly experimentally and mathematically.

pkucvcourse Computer Vision: Image Processing, Classification, Object Detection, Semantic Segmentation, etc.
Peking University, Institute for Artificial Intelligence
advisor: Prof. Yixin Zhu, Dr. Siyuan Huang
course page / problem sets / final project

Dealt with various kinds of quintessential CV tasks in the problem sets. In the final project, we used a transformer-based model to deal with transparent object segmentation problems. This task aims to develop a vision-based biological experiment monitoring system where transparent objects are pervasively used.

Accumulated experiences in image processing, OpenCV, and PyTorch-based deep learning in CV field, including image classification, object detection, semantic segmentation, and other aspects during this final project and other course projects.

modularrobotics Modular Robotics: Modular UAV & Modular WMR
Beijing Institute for General Artificial Intelligence (BIGAI), Robotics Lab
advisor: Dr. Yao Su, Dr. Meng Wang, Dr. Hangxin Liu

Explored ground- and air-based modular robotics. Used graph theory related method to determine the optimal structure of our self-built novel multi-UAV system, and pave the way for building a multifunctional agile multi-UAV system.

Designed and constructed an omni-wheeled dockable Wheel Mobile Robot (WMR). we investigated the docking issue and carried out the group transportation with our multi-WMR system.

3dforcesensor Force sensor with Gelsight: Novel Sensor design and application with manipulator
Beijing Institute for General Artificial Intelligence (BIGAI), Robotics Lab
advisor: Dr. Wanlin Li, Dr. Meng Wang, Dr. Hangxin Liu

Designed and built the L³F-TOUCH sensor, an enhanced version of the classic GelSight sensor, to acquire a much better three-axis force sensing capability while being light-weight, low-cost and wireless for the ease of replication and deployment.

Learned how to design and carried out persuasive experiments. Accumulated rich experience in hardware design and manipulator programming (sensing, planning, and actuating) with Moveit! and ROS.



Design and source code from Jon Barron's website