I am a postgraduate student supervised by Prof. Tao Zhang (IET Fellow, Head of Department) in the Institute of Navigation and Control (INC) at Tsinghua University.

Before stduying at Tsinghua, I received my B.E. in Mechatronics Engineering from University of Beijing Institute of Technology in 2019. Besides, I worked as a research assistant in Beijing Advanced Innovation Center for Intelligent Robots and Systems (BAICIRS) supervised by Prof. Qing Shi (IEEE Senior Member) and Prof. Toshio Fukuda (IEEE President(2020)).

Location: Main Building 428, Institute of Navigation and Control, Department of Automation, Tsinghua University, Beijing, China

Email: wangsj19@mails.tsinghua.edu.cn

News

• Mar. 2022, One paper is accepted by T-RO(Top Journal in Robotics).
• Feb. 2022, One paper is accepted by RA-L with ICRA2022 option.
• Sept. 2021, I confirmed to be the Session Chair, Mobile and Bimanual manipulation, at IROS2021.
• Sept. 2021, I will make an spotlight presentation at IROS2021.
• Jul. 2021, I made an oral presentation at CCC2021.
• Jul. 2019, I made an oral presentation at AIM2019 and our paper won the best student paper award.
• Jun. 2019, earned my B.S. degree as outstanding undergraduate

Research Interests

My research interests include intelligent robotics and reinforcement learning. To be specific, my research focuses on trajectory planning of free-floating space robots using reinforcement learning and developing of small-scale quadruped robots capable of multi-modal motions. To develop a general-purpose learning framework, I am interested in studying topics including, but not limited to, Reinforcement learning, Optimization, Biomimetic Robot Design, Control Theory, SLAM, Deep Learning, Dynamics Modeling and Simulation, Image Processing.

When I was at BIT, I worked with Prof. Qing Shi and Prof. Toshio Fukuda to study the intersection of robotics and biology. For example, to realize the detection and operation in narrow spaces, I developed a miniature rat-inspired robot that captures a decent representation of an actual rat, and designed a hierarchical multi-modal locomotion controller based on optimization with bionic constraints. Our robot can execute multiple modes including crouch-standing action, walking forward, turning around, crawling and recovery from a fall. This work won the best student paper award at IEEE AIM in 2019.

After pursuing my master degree, I worked with Prof. Tao Zhang on the application of deep learning on space robots. For instance, traditional model-based methods have the singular problem in trajectory planning of free-floating space robots, so I designed a multi-target trajectory planning strategy of a 6-DoF free-floating space robot using reinforcement learning. On the other hand, in order to combine the perception, planning and control into an integrated system, I designed an end-to-end trajectory planning policy for a free-floating space robot, and implemented a deep reinforcement learning algorithm based on maximum entropy to search the optimal policy.