CV

I am Jiawei Fu, an interdisciplinary student researcher on computer vision and robotics.

Table of contents

Basics

Name Jiawei Fu
Position Research Intern
Email jiawfu@student.ethz.ch
Url https://jayefu.github.io/
Summary An interdisciplinary student researcher on computer vision and robotics
Status I am actively searching for Ph.D. positions in the field of computer vision and robotics

Research

  • 2023.01 - now
    Morphology Optimization for Visual Sensor Parameters
    Advised by Andrei Atanov, Andrew Spielberg ,and Prof. Amir Zamir
    Visual Intelligence and Learning Lab, EPFL, Lausanne, Switzerland
    • Developed the computational framework for jointly optimizing visual sensor parameters and control
    • Validated the bio-inspired hypothesis about the acute zones in dragonfly eyes
    • Conducted experiments in the Habitat Simulator for synthesized environments and realistic scenes
  • 2022.04 - 2022.09
    Building Multi-sensor System for Autonomous Driving
    Advised by Prof. Dengxin Dai
    Vision for Autonomous Systems, Max Planck Institute for Informatics, Saarbrucken, Germany
    • Designed the synchronization circuit among cameras, LiDARs, GPS, and edge computing devices
    • Implemented the algorithms for calibration, synchronization, data collection, and post-processing
    • Designed the electrical and mechanical structure for binding the sensors
  • 2021.09 - 2022.03
    Learning Deep Sensorimotor Policies for Vision-based Drone Racing
    Advised by Yunlong Song, Yan Wu, Fisher Yu, Prof. Davide Scaramuzza
    Robotics and Perception Group, University of Zurich, Zurich, Switzerland
    • Developed a deep sensorimotor policy with imitation learning for vision-based drone racing
    • Enhanced the robustness of the visual encoder with contrastive learning and data augmentation
    • Conducted flight experiments in the Flightmare simulator and achieved state-of-the-art performance
  • 2021.03 - 2021.07
    Computational Design of ANYmal-on-Wheels
    Advised by Dr. Marko Bjelonic, Joonho Lee, Prof. Marco Hutter
    Robotic Systems Lab, ETH Zurich, Zurich, Switzerland
    • Developed the joint optimization framework for the leg morphology and control of ANYmal-on-Wheels
    • Used Bayesian optimization to learn morphology and reinforcement learning to optimize control
    • Conducted experiments in the RaiSim simulator and reduced the energy consumption during locomotion
  • 2019.11 - 2020.06
    Cooperation of Mobile Manipulator and Drone
    Advised by Prof. Jiwen Zhang
    Institute of Mechanical and Electronic Engineering, Tsinghua University, Beijing, China
    • Built the simulation environment for the mobile manipulator and the drone
    • Developed the pose allocation algorithm for the end effector and the drone based on safety cost
    • Implemented the inverse kinematics solving algorithm with MoveIt
  • 2018.03 - 2019.01
    Hardware Development for Humanoid Soccer Robot
    Advised by Prof. Jiwen Zhang
    Institute of Mechanical and Electronic Engineering, Tsinghua University, Beijing, China
    • Designed the circuit for connecting a micro-controller, servo motors, and IMU
    • Developed the real-time operating system for the micro-controller
    • Implemented the ball detection algorithm and the behavior decision system for humanoid robots

Work

  • 2022.04 - 2022.09
    Research Intern
    Advised by Prof. Dengxin Dai
    Topic: Building Multi-sensor System for Autonomous Driving
    Vision for Autonomous Systems, Max Planck Institute for Informatics, Saarbrucken, Germany
    • Designed the synchronization circuit among cameras, LiDARs, GPS, and edge computing devices
    • Implemented the algorithms for calibration, synchronization, data collection, and post-processing
    • Designed the electrical and mechanical structure for binding the sensors
  • 2021.09 - 2022.03
    Student Intern
    Advised by Prof. Davide Scaramuzza
    Topic: Learning Deep Sensorimotor Policies for Vision-based Drone Racing
    Robotics and Perception Group, University of Zurich, Zurich, Switzerland
    • Developed a deep sensorimotor policy with imitation learning for vision-based drone racing
    • Enhanced the robustness of the visual encoder with contrastive learning and data augmentation
    • Conducted flight experiments in the Flightmare simulator and achieved state-of-the-art performance

Education

Awards

Skills

Programming Language
Python
C++
C
MATLAB
Framework
PyTorch
TensorFlow
ROS
CMake
Simulator
Habitat
Flightmare
RaiSim
Isaac Sim
Developer Tool
Docker
Singularity
Git
Hardware
Jetson
RGBD Camera
LiDAR
GPS

Languages

Chinese
Native speaker
English
Fluent
German
Basic