Robotics and Biology Laboratory

Xing Li

Research Interests

Hi, my name is Xing Li. I am a PhD student at the Technische Universität Berlin under the supervision of Professor Oliver Brock. I am associated with the Excellence Research Cluster Science of Intelligence at Berlin. 

My reserach interest is to develop a novel system that allows non-experts to intuitively teach robots complex skills through demonstrations. Different to other reserach works focus on Learning from demonstration, I am focusing on how to endow the robot with the ability to efficiently explore the environment under the guidance of human demonstrations. I believe this is a possible solution to bridge the gap in perception repertoire between humans and robots is to leverage the potential provided by the robots’ embodiment. I hope my research could contribute to the robotics community in the context of bringing robots to everyday household environments and making them accessible to everyone.



Learning To Manipulate From Demonstration

We will develop a novel approach to Learning from Demonstration for teaching a robot complex, contact-rich manipulation tasks. This approach will enable the robots to physically operate complex locks and other mechanisms.



Li, Xing; Brock, Oliver
Learning From Demonstration Based on Environmental Constraints
IEEE Robotics and Automation Letters, 7 (4) :10938-10945


Rosbach, Sascha; James, Vinit; Großjohann, Simon; Homoceanu, Silviu; Li, Xing; Roth, Stefan
Driving Style Encoder: Situational Reward Adaptation for General-Purpose Planning in Automated Driving
2020 IEEE International Conference on Robotics and Automation (ICRA), Page 6419-6425
Rosbach, Sascha; Li, Xing; Großjohann, Simon; Homoceanu, Silviu; Roth, Stefan
Planning on the fast lane: Learning to interact using attention mechanisms in path integral inverse reinforcement learning
2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Page 5187-5193