跳过正文

About Me

331 字·2 分钟
Haoran Zhou | 周浩然
作者
Haoran Zhou | 周浩然
海阔凭鱼跃
目录

Hello! I’m Haoran Zhou, currently working as a dexterous hand algorithm engineer at Xiaomi Robotics. My research interests include reinforcement learning and dexterous manipulation.

Feel free to contact me. My email is jou072@126.com.

MEng in Mechatronics Engineering, 2025
上海理工大学, University of Shanghai for Science and Technology (USST)

Experience
#

  1. 上海市主动健康协同创新中心

    2023/3 - 2023/9

    派遣实习

  2. 上海卓易得机器人有限公司

    2023/11 - 2024/3

    机器人算法实习生

  3. 上海傅里叶智能科技有限公司

    2024/5 - 2024/8

    机器人算法实习生

  4. 北京小米机器人科技有限公司

    2024/8 - 2025/7

    机器人算法实习生

  5. 北京小米机器人科技有限公司

    2025/7 - now

    灵巧手操作算法工程师

Awards
#

  • 中国研究生机器人创新设计大赛 - 国三等奖
  • 大学生工程训练与创新能力竞赛 - 国银奖 & 省特等奖
  • 大学生机械创新竞赛 - 省一等奖
  • 大学生电子设计竞赛 - 省一等奖
  • 人工智能创新竞赛 - 省一等奖
  • Robocom 竞赛 - 省三等奖
  • 小米 - 最佳新人奖

Projects
#

NoneJou072/robopal

robopal: a multi-platform, modular robot simulation framework based on MuJoCo, mainly used for reinforcement learning and control algorithm implementation of robotic arms.

Python
152
15
NoneJou072/robochain

A simulation framework based on ROS2 and LLMs(like GPT) for robot interaction tasks in the era of large models

Python
110
12

Publications
#

Intelligent Redundant Manipulation for Long-Horizon Operations with Multiple Goal-conditioned Hierarchical Learninghomepage
Haoran Zhou, Xiankun Lin
Advanced Robitics(Journal), 2024

Abstract purple artwork
We propose a Multiple Goal-conditioned Hierarchical Learning (MGHL) framework to tackle long-horizon task manipulated by robotic arms.

DyDexHandover: Human-like Bimanual Dynamic Dexterous Handover using RGB-only Perceptionhomepage
Haoran Zhou, Yangwei You, Shuaijun Wang
Arxiv, 2025

We present DyDexHandover, a framework for dynamic bimanual handover, achieving human-like throwing and catching using RGB-only perception.

TacRefineNet: Tactile-Only Grasp Refinement Between Arbitrary In-Hand Object Poseshomepage
Shuaijun Wang, Haoran Zhou, Diyun Xiang, Yangwei You
Arxiv, 2025

we propose TacRefineNet, a tactile-only framework that achieves fine in-hand pose refinement of known objects in arbitrary target poses using multi-finger fingertip sensing.

Collaborative publications
#

The Development of a Visual Servo Simulator Based on ROS2 and MuJoCo
Yang Lu, Zongdao Li, Haoran Zhou, Liang Zhou
IEEE International Conference on Robotics and Biomimetics (ROBIO), 2023

Deep-reinforcement learning aided dynamic parameter identification of multi-joints manipulator
Zhuoran Bi, Wenlong Zhao, Yichao Huang, Haoran Zhou and Qingdu Li
International Journal of Intelligent Systems Technologies and Applications, 2024