Avatar

Yiming Li

Ph.D in Robotics

École polytechnique fédérale de Lausanne (EPFL)

About me

Hello, welcome to my website! I am currently a PhD student at Ecole Polytechnique Federale de Lausanne (EPFL) and a research assistant in the Robot Learning & Interaction group at Idiap Research Institute supervised by Dr. Sylvain Calinon. My current research interests mainly focus on robot manipulation, optimal control and robot perception.

Interests

  • Robot Manipulation
  • Optimal Control
  • Machine Learning
  • Robot Perception

Education

  • MSc in Control Theory and Control Engineering, Jun, 2022

    Institute of Automation, Chinese Academy of Sciences

  • BSc in Mechanical Engineering, Jun, 2019

    Tongji University

News

[10/2022] I start my PhD in the Robot Learning & Interaction group at Idiap Research Institute.

[08/2022] One paper accpeted by IEEE Transactions on Cognitive and Developmental Systems (T-CDS).

[02/2022] One paper accpeted by ICRA’22.

[12/2021] One paper accpeted by IEEE Robotics and Automation Letters (RA-L).

[07/2021] One paper accpeted by IROS’21.

[02/2020] I joined ByteDance AI Lab as a research intern.

Publications

Simultaneous Semantic and Collision Learning for 6-DoF Grasp Pose Estimation

Grasping in cluttered scenes has always been a great challenge for robots, due to the requirement of the ability to well understand the …

DVGG: Deep Variational Grasp Generation for Dextrous Manipulation

This work presents DVGG, an efficient grasp generation network that takes single-view observation as input and predicts high-quality …

HGC-Net: Deep Anthropomorphic Hand Grasping in Clutter

Grasping in cluttered environments is one of the most fundamental skills in robotic manipulation. Most of the current works focus on …

Projects

*

6-DoF Robotic Grasp Pose Estimation

We proposed a simultaneous network to jointly learn instance-level, collision-free 6-DoF grasp poses in cluttered.

FightingICE Game AI Competition

We developed several RL-based AIs to fight against existing opponents, and defeated MctsAi (baseline) with a win rate of over 90%.

HIT DLR hand teleoperation

We developed a real-time hand-arm teleoperation system based on ROS for imitation learning on dextrous manipulation (UR5 with HIT-DLR Hand)

Personal Urban Mobility Access (PUMA)

PACE Vehicle Engineering Center, Tongji University.

RoboMaster Competition

Super Power Robot Team, Tongji University.

Honors

[2022] Outstanding students of the Institute of Automation, Chinese Academy and Sciences

[2019] Excellent Graduates of Shanghai

[2018] First Prize in RoboMaster National College Student Robot Contest

[2018] First Prize in PACE Global Project Competition

[2018] Meritorious Winner in Mathematical Contest In Modeling

[2018] Tongji Scholarship of Excellence

[2017] Second Prize of Mathematics Competition of Chinese College Students

[2016] Tongji Scholarship of Excellence

Contact

  • ymli.cn@gmail.com
  • (+86)155 5656 1628
  • 7 Rue Pre-Borvey, Martigny, 1920