Collaborative Research on Senior Design Publication with ZJUI Alumni
Haining, China
Research Contributor, Advisor: Zihao Li (Ph.D. Candidate, UIUC)
Dec 2023 – Jan 2024
- Conducted an extensive literature review on cutting-edge Vehicle-to-Vehicle (V2V) communication technologies and their applications in intelligent transportation systems.
- Authored the 'Related Work' section, providing a comprehensive overview of current advancements and their relevance to the project.
- Contributed to manuscript refinement by paraphrasing and enhancing the paper content, ensuring alignment to academic standard and conference guideline
- Wrote, edited, proofread the paper, "Real-time Vehicle-to-Vehicle Communication Based Network Cooperative Control System through Distributed Database and Multimodal Perception: Demonstrated in Crossroads", \n
as the fifth author, which is accepted to the ICICT 2024 conference with Oral Presentation.
- On behalf of our team, I delivered the Oral Presentation on the conference.
Intelligent Vehicle Risk Warning System Integrated with Large-Language Models
Hangzhou, China
Summer Research Intern, Advisor: Prof. Wei Xiang
Jul 2023 – Sep 2023
- Developed a LLM framework for human-machine co-driving scenario using state-of-the-art computer vision models including Bytetrack, Segment Anything, and Grounding DINO, to generate potential risk analytics from monocular visual data, forming the basis for a dynamic risk warning system interface.
- Proposed a hybrid frame analysis method to interpret complex driving environment, which could estimate the intentions of pedestrians and offer real-time adaptive warnings based on multi-object tracking and eye-tracking technology.
- Refined HCI design and the eye-movement technology and risk alert design; completed a 41-person user experience experiment based on a simulated cockpit and handled the result data processing.
- Wrote, edited, proofread the manuscript the paper, "Visionary Co-Driver: LLMs Enhance Driver Risk Perception with ARHUD", as the third author, which is submitted to the CHI 2024 conference in the Computational Interaction.
Open-source Software Community Practice and Effectiveness Research
Hangzhou, China
Jun 2022 – Oct 2022
Summer Research Assistant, Advisor: Prof. Zhiyuan Wan
- Analyzed 40+ open-source projects using tools like Designite and SonarQube, identifying common architectural anti-patterns: Cyclic Dependency, Ambiguous Interface, Scattered Functionality, etc.
- Constructed datasets of 20 Java and 20 C/C++ projects, covering big data, network services, network framework and embedded systems for tool efficacy analysis.
- Mapped version history against source code evolution to assess architectural improvement and technique debt reduction, benchmarking tool efficiency analysis.
- Designed an academic poster “Architectural Antipatterns Research Based on Software Engineering”, wining first prize for ZJU undergraduate summer research competition.
Our acedemic poster won first prize in undergraduate group. Check it here!
ZJU SRTP, Simulation and modeling of unmanned vehicle safety test based on Apollo D-kit
Haining, China
Mar 2021 – May 2022
SRTP Assistant, Advisor: Prof. Simon Hu
- Assemble and config the Baidu Apollo D-kit Self-driving Test Car, including Lidar, Radar and Camera module.
- Used Python to process Waymo dataset, data cleaning and RNN network construct to gather the real-road statistic information, analyzed and recorded the common traffic states of unmanned vehicles.
- Constructed the high-solution Lidar map of ZJU international campus via Apollo D-kit Self-driving Test Car.
- Proposed a perception range model for Connected and Automated Vehicles(CAVs) in traffic state estimation based on recorded data and literature review
My SRTP image is shorten as the poster for ZJU 2022 Summer School!
