Renyuan Liu

+86 14784206312 | rliu@e.gzhu.edu.cn

Education

• Guangzhou University
  Sept. 2022 - Jun. 2026 (Expected)
  B.Eng. in Computer Science (Information Security); GPA: 89.81/100.00; Ranking: Top 10%
  Curriculum: Machine Learning 100*, Data Structure and Algorithm Laboratory 99*, Operating System 98* (Course Project 95*), Programming Practice 98*, Data Structure and Algorithm 97*, Programming Laboratory I 95*, Computer Network (Course Project 95*), Principles of Computer Composition, Higher Mathematics, Discrete Mathematics, Linear Algebra, Probability and Mathematical Statistics (*: rank 1st in all students of the course).

• The University of Hong Kong/University of Macau (Summer Camp)
  Nov. 2023
  GPA: 97.50/100.00 (Interdisciplinary Programme)
  Honor: Commendation Letter for Outstanding Performance in the Winning Team

Manuscripts Under Review

• R. Liu, H. Zhou, C. Fang and Q. Fu, Manuscript under double-blind review. Submitted to International Conference on Robotics and Automation (ICRA)(CCF-B).

• M. Wang^*, R. Liu^*, W. He, and Q. Fu, A Neuronal Assembly Model with Elevated Time Derivative Boosts Loom-Selectivity. Submitted to PeerJ Computer Science (JCR Q2).

Publications

• R. Liu and Q. Fu, Attention-Driven LPLC2 Neural Ensemble Model for Multi-Target Looming Detection and Localization. The 2025 International Joint Conference on Neural Networks (CCF-C, acceptance rate ≈ 38%).

• G. Gao^*, R. Liu, M. Wang and Q. Fu^*, A Computationally Efficient Neuronal Model for Collision Detection With Contrast Polarity-Specific Feed-Forward Inhibition. Biomimetics, vol. 9, no. 11, p. 650, 2024 (JCR Q1, IF = 3.4).

• C. Fang^*, H. Zhou, R. Liu, and Q. Fu^*, A neuromorphic binocular framework fusing directional and depth motion cues towards precise collision prediction. Neurocomputing, 131660 (JCR Q1, IF = 6.5).

• H. Zhou, C. Fang, R. Liu, and Q. Fu, A Bio-Plausible Neural Network Integrating Motion and Disparity Pathways for Looming Perception. Acta Electronica Sinica, p.1-16, 2025 (CCF-A in Chinese Category).

• J. Huang^*, Z. Qin, M. Wang, R Liu, and Q. Fu^*, A Biomimetic Collision Detection Visual Neural Model Coordinating Self-and-Lateral Inhibitions. The 14th International Conference on Biomimetic and Biohybrid Systems (Living Machines 2025)(Oral).

Honors and Awards

• First Prize (Provincial; Top 3%), Chinese Collegiate Computing Competition (4C)
  May 2025

• Honorable Mention (International), Mathematical Contest in Modeling (MCM)
  Jan. 2025

• First Prize (National; Top 5%), Asia and Pacific Mathematical Contest in Modeling (APMCM)
  Nov. 2024

• First Prize & Innovation Silver Award (Provincial; Top 2 out of 1,167 Teams),
  “Greater Bay Area Cup” Guangdong-Hong Kong-Macao Financial Mathematics Modeling Competition
  Nov. 2024

• Third-Class Scholarship (Top 12%), Guangzhou University
  Nov. 2024

• First-Class Scholarship (Top 5%), Guangzhou University
  Nov. 2023

Research Experience

Computational Autonomous Learning Systems Lab  Advisor: Prof. Pengcheng Liu
Department of Computer Science, University of York, York, UK (On-Site)

• Bio-inspired models and biologically-plausible mechanisms for life-long learning
  Jun. 2025 – Sept. 2025

  • Robotic arm motion planning: Learned expert-guided trajectory optimization via Learning from Demonstration (LfD) and applied biologically inspired probabilistic movement primitives (ProMPs) for push–grasping with the Franka Emika Panda.

  • Navigation and manipulation: Working on developing a lifelong learning navigation–manipulation system on TurtleBot 3 with OpenMANIPULATOR-X that adapts to new environments while retaining performance in previously learned ones.

Machine Life and Intelligence Research Centre  Advisor: Prof. Qinbing Fu
School of Mathematics and Information Science, Guangzhou University, Guangzhou, China

• Real-time Visual Processing Systems Development of Micro-Mobile Robot
  Mar. 2023- Present
  • Reading and giving reports of research articles during research seminars on a weekly basis.
  • Deployed visual neural network models inspired by insect neurons onto the STM32-based micro-robot Colias, achieving real-time collision perception and avoidance. Optimized model memory usage to fit within the 62 KByte SRAM capacity of Colias; developed and refined algorithms to enable real-time execution under extreme computational constraints (processing time < 33 ms on the STM32F427 chip); performed debugging, tuning, and conducted both offline and online experiments.
  • Fly-Inspired Ultra-selective Looming Perception and Avoidance on Resource-Constrained Micro-Robots, poster presentation at the 26th Towards Autonomous Robotic Systems (TAROS 2025).
  • Selected code can be accessed below:
    Fly Visuomotor-Inspired Attention-LPLC2 Model (independently, 2k lines of code in C);
    Locust Vision-Inspired Optimized-LGMD Model (independently, 1k lines of code in C).
• Attention-Driven LPLC2 Neural Ensemble Model for Multi-Target Looming Detection and Localization, paper accepted at IJCNN 2025, first author.
  Jul. 2024 - Nov. 2024
  • Conducted full-cycle research on modeling the lobula plate/lobula columnar type 2 (LPLC2) neural ensemble in the fruit fly Drosophila, known for its ultra-selectivity to looming stimuli.
  • Developed the multi-attention LPLC2 (mLPLC2) neural network model inspired by the visual system of the fly by leveraging a bottom-up attention mechanism driven by motion-sensitive neural pathways (independently, 3k lines of code in C/C++).
• A Computationally Efficient Neuronal Model for Collision Detection with Contrast Polarity-Specific Feed-Forward Inhibition, article published at Biomimetics, second author.
  Mar. 2024 - Jul. 2024
  • Participated in the entire research on modeling the optimized locust lobula giant movement detector neuron with detailed feed-forward inhibition (oLGMD) to enhance processing speed and the robustness towards translating movement.
  • Implemented oLGMD model into the embedded system of Colias, and conducted closed-loop arena comparative experiments to evaluate performance of oLGMD, achieving the highest success ratio of collision avoidance at 97.51% while nearly halving the processing time compared with previous LGMD models; conducted all online experiments of this paper, analyzing the results using real-world data collected by the Colias robot; designed criteria to assess time efficiency and collision selectivity.
  • Led the initial writing of the introduction and experimentation sections; participated in revising the submitted paper.
• Bio-Inspired LGMD Collision Detection Model Leveraging Optical Flow and Learning-Based Optimization, Provincial Key College Students’ Innovative Entrepreneurial Training Plan Program.
  Mar. 2023 - Present
  • Developed neuromorphic binocular models for collision prediction which combines directional and depth motion cues; optimized directional-selective neuron parameters using a genetic algorithm; collected a stereo RGB-D dataset capturing diverse indoor-outdoor collision scenarios to support model training and evaluation; conducted online robotic experiments with the TurtleBot 4 robot.
  • Designed detailed figures illustrating the models and experiments; drafted manuscript introductions, and contributed to manuscript revisions.

Skills / Learning is the one of the happniest thing in the world

• Language: IELTS 6.5 (R8.0, L6.5, W6.0, S5.5), CET-6 564 (242/248.5 in the reading section)
• Programming Skills: C/C++, Python, Matlab
• Others: ROS, LaTeX, Keil, Webots, Linux, Git, Markdown, MS Office/Visio, Adobe Photoshop/Premiere Pro

Hobbies: Movie, Music, Photography, Basketball, Jogging, Badminton, Hiking, Cooking.