Yifan Peng
Yifan Peng
Forge ahead, overcoming any obstacles along the way | 逢山开路，遇水搭桥

⭐ Biography

Hi 😊! I am a third-year postgradute student at School of Automation, Nanjing University of Science and Technology (NJUST, 211 project). My major is Control Science and Engineering, and my research focuses on mobile robot localization and scene understanding.

I received my Bachelor’s degree in Automation at Nanjing University of Science and Technology (NJUST) in June 2021, as a member of Robotics Innovation Lab (Auto-Robot).

Academic performance:

  • Master’s: Average score: 91.48/100，ranked 11/390 (top 3%).
  • Bachelor’s: GPA 3.42/4, ranked 56/188 (top 30%).


🚀 Research Interests

My research interests focus on:

  • Deep Learning-based Visual SLAM in dynamic scene
  • Robot environment perception and understanding
  • Multi-robot collaborative SLAM

Currently, SLAM remains a fundamental task in robotics. In the future, I hope to focus on the integration of deep learning techniques (such as generative models and semantic segmentation, etc.) with robotics SLAM, enabling them to perceive and understand the world like humans do, so that they can be truly applied in human life and work collaboratively with people in the future.


📜 Publications

Paper:

[1] Yifan Peng, Rui Xv, Yvmeng Xv, Yifei Wu and Qingwei Chen. A High-Precision Dynamic RGB-D SLAM Algorithm for Environments with Potential Semantic Segmentation Network Failures. (IEEE Transactions on Instrumentation & Measurement, JCR-Q1, Under review).

[2] Yifan Peng, Rui Xv, Yvmeng Xv and Yifei Wu. A Visual SLAM Method Based on Semantic Segmentation Network in Dynamic Scenes. (Accepted at the 22nd IEEE International Conference on Industrial Informatics (IEEE-INDIN 2024))

Patents：

[1] Yifei Wu, Jialun Yu, Jian Guo, Qingwei Chen, Yvmeng Xv, Yifan Peng, Rui Xv and Qiwei Ma. A Wheel-Legged Air-Ground Integrated Reconnaissance Robot Based On Super-Spiral Sliding Form.

Software copyrights：

[1] A Warehouse Management System V1.0 Based on Multi-AGV V1.0.

[2] MATLAB-Based Multi-AGV Warehouse Management System Algorithm Simulation Program V1.0.


📡 Project Experience

Here are projects I participated in during my undergraduate and master’s studies. More details and outcomes of all these projects are shown in Portfolio page.

During my master’s and undergraduate studies, I participated in the following projects closely related to my research interests. These projects involved visual SLAM, laser SLAM, and the Robot Operating System (ROS):

  1. Research on Visual SLAM Algorithms for Mobile Robots in Complex Dynamic Environments. (2024.02 ~ present)
  2. Design of a Navigation and Localization System for an Autonomous Harvesting Robot in a Simulated Farmland Environment. (2023.09 ~ 2023.11)
  3. Design of a Laser SLAM System Based on LeGO-LOAM. (2023.05 ~ 2023.06)
  4. Design of an Autonomous Vehicle System Based on ROS for Multi-Vehicle Interaction Scenarios. (2022.09 ~ 2022.11)
  5. Design of a Navigation and Localization System for a Quadruped Bionic Robot. (2021.01 ~ 2021.06)

Additionally, I have participated in the following projects, which required not only theoretical or engineering design but also the writing of related project proposals, patents, and theses:：

  1. Research on Integrated Methods for Scene Understanding and Knowledge Acquisition and Reasoning in Earth-Moon Space Based on Generative Models. (2023.09 ~ present)
  2. Design of a Control System for a Wheeled-Legged Amphibious Robot Based on Super-Helical Sliding Mode Control. (2023.02 ~ 2023.03)
  3. Design and Implementation of a Multi-AGV Warehouse Management System. (2019.10 ~ 2020.05)

During my undergraduate studies, I also participated in the following competition-related projects:

  1. Design of an Intelligent Car Based on Visual Guidance. (2019.03 ~ 2019.07)
  2. Design of an Electromagnetic Railgun Based on Visual Guidance. (2019.08 ~ 2019.10)
  3. Design of a Vehicle-Type Exploration and Cruising Robot. (2019.10 ~ 2019.12)


🔥 Skills

1.Visual SLAM:

  • Theories: feature extraction and matching, pose estimation and optimization, pointcloud map creation.
  • Learning-based Dynamic Visual SLAM Algorithms: especially detection and handling of dynamic targets.
  • Open-source Systems: ORB-SLAM2/3, DS-SLAM, SG-SLAM (code reading and secondary development).

2.Programming Development:

  • Language: C/C++; Python (a little).
  • Platform: Ubuntu (frequently used command line; code writing and debugging).

3.Tools: ROS (distributed communication framework); OpenCV (image processing); OpenMMlab (open-source deep learning framework); G2O (optimization); PCL (point cloud mapping); GitHub (code management).

4.Engineering practice:

  • PCB hardware design.
  • Proficient in using common laboratory equipment (soldering iron, multimeter, oscilloscope, etc.).
  • Capable of independently designing and building robots.

5.Software: MATLAB、CLion、Altium Designer、Keil 5、Photoshop、Word、PowerPoint, etc.

6.Research and Writing Abilities: Proficient in academic reading, writing, drawing, and literature research.


📧 Contact Information

Please feel free to contact me at following ways:

  • Email : effunwayj@njust.edu.cn
  • Wechat : pyf140718
  • Phone number : +86 18852056252
  • Facebook : YiFan Peng
  • Personal website : effun.xyz