Yuanpei Chen | 陈源培

I am a fourth-year undergraduate student at South China University of Technology, working with Prof. Yaodong Yang at Peking University. I also had the privilege of working closely with Prof. Xiaolong Wang at UCSD, Prof. Chenguang Yang at SCUT and Prof. Hao Dong at Peking University. In 2023, I am fortunately working as a visiting researcher at Stanford University, advised by Prof. Karen Liu and Prof. Fei-Fei Li.

Email  /  CV  /  Google Scholar  /  Github /  WeChat

I'm interested in robotics, dexterous manipulation, reinforcement learning, and sim-to-real transfer. My goal is to let the robot perform various cool behaviors in the real world as in the simulation.

We introduce a hierarchical framework that uses human hand motion data and deep reinforcement learning to train dexterous robot hands for effective object-centric manipulation in both simulation and real world.

We present Sequential Dexterity, a general system based on reinforcement learning (RL) that chains multiple dexterous policies for achieving long-horizon task goals.

We design a system with two multi-finger hands attached to robot arms to solve the dynamic handover problem. We train our system using Multi-Agent Reinforcement Learning in simulation and perform Sim2Real transfer to deploy on the real robots.

We propose a bimanual dexterous manipulation benchmark (Bi-DexHands) according to literature from cognitive science for comprehensive reinforcement learning research.

We propose Maniwhere, a generalizable framework tailored for visual reinforcement learning, enabling the trained robot policies to generalize across a combination of multiple visual disturbance types.

We propose the neural attention field for representing semantic-aware dense feature fields in the 3D space by modeling inter-point relevance instead of individual point features.

We present GarmentLab, a content-rich benchmark and realistic simulation designed for deformable object and garment manipulation.

We introduce ReDMan, an open-source simulation platform that provides a standardized implementation of safe RL algorithms for Reliable Dexterous Manipulation.

We propose a framework to learn a universal human prior using direct human preference feedback over videos, for efficiently tuning the RL policy on 20 dual-hand robot manipulation tasks in simulation, without a single human demonstration

Reconfiguring a die to match desired goal orientations. This task require delicate coordination of various muscles to manipulate the die without dropping it.

We investigate safe MARL for multi-robot control on cooperative tasks, in which each individual robot has to not only meet its own safety constraints while maximising their reward, but also consider those of others to guarantee safe team behaviours.

In this study, we take advantage of visual affordance by using the contact information generated during the RL training process to predict contact maps of interest.

We propose Simulation Twin (SimTwin) : a deep reinforcement learning framework that can help directly transfer the model from simulation to reality without any real-world training.

I was quite into competitive robot🤖 and used to compete in RoboMaster🏅 and ICRA 2021 AI Challenge.

Template stolen from Jon Barron. Thanks for stopping by :)
Last updated: Nov 6, 2021