Zewei Zhang

A master's student at EPFL, working on robotics.

prof_zewei.jpeg

I am a Master’s student in the Department of Mechanical Engineering at Ecole Polytechnique Fédérale de Lausanne. During my master s study, I have the privilege of conducting the research projects, in Biorobotics Laboratory (BioRob) at EPFL advised by Prof. Auke Jan Ispeert and in Robotics System Laboratory (RSL) at ETH Zurich advised by Prof. Marco Hutter. My research interests lie in the field of robot learning, particularly in locomotion control using reinforcement learning or bio-inspired control. My dream is to create a robot that can help humanity revolutionize the production and explore the universe. Prior to joining EPFL, I completed my bachelor’s degree at Waseda University in Tokyo, Japan, where I focused on space exploration.

During my spare time, I love to play basketball and enjoy the hotpot :).



Publications


sym

Motion Priors Reimagined: Adapting Flat-Terrain Skills for Complex Quadruped Mobility
Zewei Zhang, Chenhao Li, Takahiro Miki, Marco Hutter
CoRL 2025

A hierarchical reinforcement learning pipeline for perceptive quadruped locomotion learning, which extracts animal motion priors learned from mocap data and generalizes to goal-reaching locomotion tasks on various types of terrains.

sym

Online Optimization of Central Pattern Generators for Quadruped Locomotion
Zewei Zhang, Guillaume Bellergarda, Milad Shafiee, Auke Ijspeert
IROS 2024 (Oral)

A framework for the online optimization of the CPG parameters through Bayesian Optimization, which can optimize and rapidly adapt to varying velocity commands and changes in the terrain.


Selected Course Projects


sym

Deep Reinforcement Learning for Quadruped Locomotion with Central Pattern Generators
Instructed by Dr. Guillaume Bellergarda, and Prof. Dr. Auke Ijspeert

A RL-based locomotion controller for quadruped robots integrating central pattern generators with foot sensory feedback, which is capable of walking on uneven terrains in simulation.

sym

Navigation in cluttered environment on quadrotor drone with onboard sensors
Team members: Chengkun Li, Zewei Zhang, Jiarui Yu, Jiale Fan, Shulang Shen,
Instructed by Dr. Shushuai Liu and Prof. Dr. Dario Floreano

Develop a navigation system for a quadrotor to maneuver through cluttered environments using onboard sensors.

sym

A Chatbot for Commonsense Persona-grounded Dialogue Generation
Team members: Zewei Zhang, Yihan Wang,
Supervised by Silin Gao, Prof. Dr. Martin Jaggi and Prof. Dr. Nicolas Flammarion

A chatbot model generating dialogue responses that possess both persona consistency and contextual coherence.