Skip to main content

Chengguang Xu

PhD Student

Contact

Office Location

440 Huntington Avenue
310 West Village H
Boston, MA 02115

Education

  • MS Control Theory and Engineering, Nankai University, China
  • BEng in Automation, Nankai University, China

About Me

  • Hometown: China
  • Field of Study: Reinforcement Learning and Computer Vision
  • PhD Advisors: Chris Amato and Lawson Wong

Biography

Chengguang Xu is a PhD student at Northeastern University in the Khoury College of Computer Sciences, co-advised by Professor Chris Amato and Professor Lawson Wong. He earned his Bachelor of Engineering and Master of Science from Nankai University in China. His main fields of research are Reinforcement Learning and Computer Vision. Specifically, he is interested in learning and planning under uncertainty by leveraging model-free deep reinforcement learning algorithms and model-based planning methods.

What are the specifics of your graduate education (thus far)?

Currently, I am working with Prof. Chris Amato and Prof. Lawson Wong on reinforcement learning problems with applications on human-robot manufacturing and vision navigation domains.

What are your research interests?

After a quite comprehensive exploration in computer vision, machine learning, and robotics,  I gradually formed my research interests in vision-based reinforcement learning methods with applications on robots.

What aspect of what you do is most interesting?

Currently, I am working on two robotic domains. One is the human-robot manufacturing domain, where the robot has to maintain estimations over the human operators and makes decisions in a partially observable environment (POMDP). The interesting parts of this domain are how to represent the behavior of the human operators and how to maintain a good estimation of human operators.  The other is the vision navigation domain, where the interesting part is how to navigate the agent in an unknown environment with limit external information.

What are your research or career goals, going forward?

Briefly speaking, being an independent researcher in my interesting research domains is my current goal.