Jingxi Xu 许靖西

jxu at cs dot columbia dot edu

PhD Student, Department of Computer Science
Member, Robotic Manipulation and Mobility (ROAM) Lab
Member, Columbia Artificial Intelligence and Robotics (CAIR) Lab
Columbia University

Visiting Student, GRASP Lab
University of Pennsylvania

Google Scholar / GitHub / Twitter

I am a second-year PhD student in Computer Science at Columbia University, co-advised by Professor Matei Ciocarlie and Professor Shuran Song. I received a Bachelor's degree from Edinburgh. I also work closely with Dinesh Jayaraman and Nikolai Matni at University of Pennsylvania. Prior to joining Columbia, I also spent some time at MIT, with Leslie Pack Kaelbling and Tomás Lozano-Pérez (2018); Columbia with Peter Allen (2018-2019). I am interested in using machine learning and vision-based methods for better planning, perception and control abilities in autonomous robots.

During my non-existent free time, I enjoy swimming, snowboarding, billiards, basketball, hiking, travelling, web-design and photography (I shoot with my Sony A7IV with the FE 25-105mm F/4 lens, FE 20mm F/1.8 lens or GoPro HERO9).

Highlights

How to grasp moving objects?

Help stroke patients reuse their hand.

Learning a tactile exploration policy.

News

[2022-03] Our work on developing a hand orthosis to help stroke patients reuse their hands is now featured on the latest issue of Columbia Engineering magazine: Better Living with Robots.
[2022-03] My guest lecture The Practice of Robot Learning is now on YouTube.
[2022-02] New preprint on tactile exploration is available.
[2022-01] One paper accepted to ICRA 2022.
[2021-12] We released our dynamic grasping code. Check it out if you want some dynamic manipulation tasks to benchmark your algorithms.
[2021-06] One paper accepted to IROS 2021.
[2021-03] One paper accepted to L4DC 2021.
[2019-04] I am deeply honored to receive the Andrew P. Kosoresow Memorial Award for Excellence in Teaching and Service from the Computer Science Department.
[2018-08] I received the CA Fellowship from Columbia University.
[2017-06] I started an internship at arm, Cambridge, UK, working on CPU pipeline visualization.
[2017-05] I graduated from the University of Edinburgh with First Class Honours and Class Medal (overall grade ranks first in class).

Publications

*: indicating equal contribution or alphabetic ordering. You can also check my Google Scholar profile.

Preprint

TANDEM: Learning Joint Exploration and Decision Making with Tactile Sensors

Jingxi Xu, Shuran Song, Matei Ciocarlie

Preprint

[arXiv] [Webpage]

Thumb Assistance Via Active and Passive Exotendons in a Robotic Hand Orthosis for Stroke

Ava Chen, Lauren Winterbottom, Sangwoo Park, Jingxi Xu, Dawn Nilsen, Joel Stein, Matei Ciocarlie

Preprint

[arXiv]

2022

Adaptive Semi-Supervised Intent Inferral to Control a Powered Hand Orthosis for Stroke

Jingxi Xu, Cassie Meeker, Ava Chen, Lauren Winterbottom, Michaela Fraser, Sangwoo Park, Lynne M. Weber, Mitchell Miya, Dawn Nilsen, Joel Stein, Matei Ciocarlie

International Conference on Robotics and Automation (ICRA) 2022

[arXiv] [Webpage]

2021

Dynamic Grasping with Reachability and Motion Awareness

Iretiayo Akinola*, Jingxi Xu*, Shuran Song, Peter Allen

International Conference on Intelligent Robots and Systems (IROS) 2021

[arXiv] [Webpage]
How Are Learned Perception-Based Controllers Impacted by the Limits of Robust Control?

Jingxi Xu, Bruce Lee, Nikolai Matni, Dinesh Jayaraman

Learning for Dynamics & Control Conference (L4DC) 2021

[arXiv] [Webpage] [Poster]

2020

Learning a Decentralized Multi-Arm Motion Planner

Huy Ha, Jingxi Xu, Shuran Song

Conference on Robot Learning (CoRL) 2020

[arXiv] [Webpage]
Learning Your Way Without Map or Compass: Panoramic Target Driven Visual Navigation

David Watkins*, Jingxi Xu*, Nicholas Waytowich, Peter Allen

International Conference on Intelligent Robots and Systems (IROS) 2020

[arXiv] [Webpage]
Accelerated Robot Learning via Human Brain Signals

Iretiayo Akinola*, Zizhao Wang*, Junyao Shi, Xiaomin He, Jingxi Xu, David Watkins, Peter Allen

International Conference on Robotics and Automation (ICRA) 2020

[arXiv] [Webpage]

2019

Active Multitask Learning with Committees

Jingxi Xu, Tang Da, Tony Jebara

🏆 Travel Award

Adaptive and Multi-Task Learning Workshop at the International Conference on Machine Learning (ICML) 2019

[arXiv] [Poster]

Teaching

Teaching assistant at Columbia University for:

Robot Learning (MECE6616) - Spring 2021
Computational Aspects of Robotics (COMS4733) - Fall 2019
Humanoid Robots (COMS6731) - Spring 2019
Computational Aspects of Robotics (COMS4733) - Fall 2018
Machine Learning (COMS4771) - Spring 2018

Talks

Tactile Sensing and Its Applications
- Guest Lecture at course Computational Aspects of Robotics (COMS4733), Columbia University, 02/25/2022
The Practice of Robot Learning (video)
- Guest Lecture at course Robot Learning (MECE6616), Columbia University, 02/08/2022
MyHand: a Wearable Hand Orthosis for Stroke
- Invited talk at Challenges and Opportunities of Human-robot Symbiosis: from Wearable Robots to Neurorobotics Workshop, IROS 2021, 10/01/2021 (Co-give with Ava Chen, Matei Ciocarlie, video)
Dynamic Grasping with Reachability and Motion Awareness

Invited talk at AIRLab, Lehigh University, 09/27/2021 (video)
Invited talk at Professor Dinesh Jayaraman's group, University of Pennsylvania, 07/14/2021
Invited talk at Professor Yi Ma's group, University of California Berkeley, 04/09/2021

ROS Tutorial

An introductory tutorial on ROS and use of robotics in the Columbia Robotics Lab, 02/08/2019

Introduction to Robotics

An invited lecture on robotics to summer school students at MIT, 08/21/2018

Projects

Deep Temporal Multi-scale Net for Apparent Personality Analysis - advised by Shih-Fu Chang
Integrate Eye-tracking with BCI System for Assistive Robots - advised by Peter Allen
Hybrid HW/SW CPU Simulation using Zync - advised by Nigel Topham and Boris Grot

Misc

I used to work on computer architecture during my undergraduate years at Edinburgh. I got accepted to DPhil in Computer Science at Oxford University with full funding (Jason Hu Scholarship from Balliol College) when I graduated from Edinburgh. I turned it down because I wanted to explore AI/robotics a bit more than staying in computer architecture for my PhD. Thus, I decided to do a self-funded master in the US to gain more research experience in AI/robotics.
I am a supporter of and allergic to 500-pound websites. I am consistently making efforts to make my website as simple as possible while still being responsive on different devices. This website is built with a single css and html file.
I am a supporter of Slow Science.
I am originally from Zhuhai, Guangdong, China. I understand Cantonese.