Boshi An | 安博施

Email  /  Google Scholar  /  Github  /  Linkedin  /  WeChat

Introduction

I am currently pursuing a master's degree at EPFL and joining the Soft Robotics Lab at ETH as a master's student, supervised by Prof. Robert Katzschmann. I received my honours degree (summa cum laude) from Turing class at the School of Computer Science, Peking University, where I was mainly advised by Prof. Hao Dong on robotics learning. I also collaborated closely with Prof. Zongqing Lu on the spatial cognitive abilities of machine learning models.

Outside of academics, I spend much of my free time at the piano, particularly in classical chamber music. During my undergraduate years I formed a piano trio with friends. Currently I am collaborating with local soprano Liudmila Arno, actively performing in Lausanne, Switzerland.

profile photo
Research

I'm interested in reinforcement learning, robotics and computational neurology.

To me, the essence of robotics lies in agility and dexterity.

Across evolution, animals have refined motor control in a billion-year interplay of body and nervous system — through a vast distributed training process governed by survival and death, extending three orders of magnitude beyond humanity’s brief venture into language, logic, and mathematics, the very emblems of our civilization. The architecture of the brain itself inscribes this history: to motion are entrusted far more neurons than ever to words.

We take pride in having taught machines to mirror our language, yet in doing so we are reminded of our limits: for what we prize as the summit of intellect, evolution regards as a late ornament.

As humble pilgrims we walk, striving to grasp nature’s vast bequest in motor control, to emulate its mastery, and perhaps one day, to venture beyond it.

Publications



Arnold: A Generalist Muscle Transformer Policy

Boshi An*, Alberto Silvio Chiappa*, Merkourios Simos, Chengkun Li, Alexander Mathis
ArXiv / Project Page / Code

Arnold is a single generalist muscle control policy trained on 14 different muscle control tasks, with 4 different embodiments. We used parallel on-policy behavior cloning, RL fine-tuning and self-distillation to match or even surpass expert performance on all tasks.




RoboVerse: Towards a Unified Platform, Dataset and Benchmark for Scalable and Generalizable Robot Learning

Co-first author among 38 authors
Project Page / Code
RSS 2025

RoboVerse is a unified framework for robotic learning. It provides an API interface to run simulations with multiple back-ends, equipped with large-scale datasets and benchmarks for training and evaluating robotic learning algorithms.




RGBManip: Monocular Image-based Robotic Manipulation through Active Object Pose Estimation

Boshi An*, Yiran Geng*, Kai Chen*, Xiaoqi Li, Qi Dou, Hao Dong
ArXiv / Project Page / Code
ICRA 2024 Oral

We achieved state-of-the-art manipulation performance by combining reinforcement learning and multi-view pose estimation.




Learning Multi-object Positional Relations via Emergent Communication

Yicheng Feng*, Boshi An*, Zongqing Lu
ArXiv / Project Page (Coming Soon) / Code (Coming Soon)
AAAI 2024 Oral

We investigated whether a compositional language for describing 2-D positional relation can emerge from communication under environmental pressure.




Bilateral Propagation Network for Depth Completion

Jie Tang, Fei-Peng Tian, Boshi An, Jian Li, Ping Tan
ArXiv / Project Demo / Code
CVPR 2024

We proposed an image-guided depth completion model which is SOTA on NYUv2 dataset and won the first place on KITTI benchmark at the time of submission.





Learning contact-rich manipulation using a musculoskeletal hand
Vittorio Caggiano*, Guillaume Durandau*, ..., Boshi An ..., Vikash Kumar
Paper
PMLR, 2023

This is the summary publication to the MyoChallenge 2022, a competition on die reorientation and baoding balls.




End-to-End Affordance Learning for Robotic Manipulation

Yiran Geng*, Boshi An*, Haoran Geng, Yuanpei Chen, Yaodong Yang, Hao Dong
ArXiv / Project Page / Code
ICRA 2023

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.





MyoChallenge: Die reorientation
Boshi An, Yiran Geng, Yifan Zhong, Jiaming Ji, Yuanpei Chen
Challenge Page / Code / Slides / Talk
First Place in NeurIPS 2022 Challenge Track (1st in 340 submissions from 40 teams)

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

Preprints



ImageManip Image-based Robotic Manipulation with Affordance-guided Next View Selection

Xiaoqi Li, Yanzi Wang, Yan Zhao, Yaroslav Ponomarenko, Qianxu Wang, Haoran Lu, Boshi An, Jiaming Liu, Hao Dong
ArXiv

We propose a framework that utilizes only RGB inputs and enables 3D articulated object manipulation.

Experience
Soft Robotics Lab, ETH Zurich, Switzerland
2025.01 - Now
Master Student in Robotics
Research Advisor: Prof. Robert Katzschmann
EPFL, Switzerland
2024.09 - Now
Master Student in Robotics
Turing Class, Peking University, China
2020.09 - 2024.07
Undergraduate Student, Summa cum laude
Research Advisor: Prof. Hao Dong
Adaptive Motor Control Lab, EPFL, Switzerland
2023.07 - 2023.08
Research Intern
Research Advisor: Prof. Alexander Mathis
University of Edinburgh, United Kingdom
2023.09 - 2023.12
Exchange Student


Selected Awards and Honors

  • 2024: Honours Degree in Science
  • 2024: Outstanding Graduates Prize (Peking University)
  • 2023: Distinguished Researcher Prize (Peking University)
  • 2022: First Place in NeurIPS 2022 Challenge Track 🏅
  • 2022: John Hopcroft Scholarship
  • 2020: Outstanding Freshman Scholarship
  • 2019: Chinese Olympiad in Informatics (NOI) Gold Medalist and National Training Team member 🏅

  • Misc

    You can find my music recordings in my Bilibili page .


    This template is a modification to Jon Barron's website.