PhD Candidate · Vehicle Engineering

Xinyu Wang

I work on end-to-end autonomous driving, multi-modal perception, and vision-language models for intelligent vehicles. Currently a visiting PhD student at the University of Glasgow.

Hefei University of Technology Visiting · University of Glasgow IEEE Graduate Student Member
Email ORCID GitHub CV (PDF)
Portrait of Xinyu Wang
Hefei · Glasgow · 2026
§ 01 / About

A short bio

Biography

I am a PhD candidate in Vehicle Engineering at the School of Automotive and Transportation Engineering, Hefei University of Technology (HFUT). From November 2025 to November 2026, I am a visiting PhD student at the James Watt School of Engineering, University of Glasgow.

My research sits at the intersection of intelligent vehicles and modern machine learning — building perception, planning and control systems that connect raw multi-sensor data to driving decisions, and increasingly, exploring how vision-language models and multi-agent systems can make these decisions more transparent and trustworthy.

I was admitted to HFUT's combined M.S.-Ph.D. program in 2021, and have worked at the Anhui Intelligent Vehicle Technology Laboratory throughout. I am a Graduate Student Member of the IEEE Robotics and Automation Society and Young Professionals.

§ 02 / Research

Four lines of inquiry

01

End-to-End Autonomous Driving

Learning-based driving stacks that map raw sensor streams directly to control, with a focus on interpretability and safe behaviour in long-tail scenarios.

02

Multi-modal Perception

Fusion of LiDAR, camera and IMU for 3D object detection — including cross-modal transformer fusion, sensor co-calibration, and bird's-eye-view representations.

03

Planning & Control

Path planning and longitudinal control for articulated vehicles and unmanned mining machines; smoothing, kinematic constraints, and real-vehicle deployment.

04

VLMs & Agents

Vision-language models and role-specialised multi-agent systems applied to driving and broader decision-making tasks. An emerging direction.

§ 03 / Publications

Selected papers

Journal Articles
2024

AFMCT: Adaptive Fusion Module Based on Cross-Modal Transformer Block for 3D Object Detection

Bingli Zhang, Yixin Wang, Chengbiao Zhang, Junzhao Jiang, Zehao Pan, Jin Cheng, Yangyang Zhang, Xinyu Wang, Chenglei Yang, Yanhui Wang
Machine Vision and Applications, vol. 35, no. 40
SCI Co-author · first author is the supervisor
DOI ↗
2024

Path Planning for Unmanned Load–Haul–Dump Machines Based on a VHF_A* Algorithm

Bingli Zhang, Yangyan Zhang, et al., Xinyu Wang
Proc. IMechE Part C: Journal of Mechanical Engineering Science
SCI Co-author · first author is the supervisor
DOI ↗
2024

Research on Single-Stage Multi-Task YOLO-Parking Algorithm Based on Obstacle and Parking Space Detection

Bingli Zhang, Yanhui Wang, Zehao Pan, Yixin Wang, Chenglei Yang, Xinyu Wang
Journal of Hefei University of Technology (Natural Science), 47(1)
EI Co-author · first author is the supervisor
DOI ↗
Conference Papers
2026

AffectMind: Proactive Knowledge Grounding with Affective Multimodal Signals for Aligned Marketing Dialogue

Xinyu Wang, Xiaomin Zhao, Yifei Kang, Zhihao Lin, Xiang Luo, Zhang Chengbiao, Jin Cheng, Yixin Wang, Yangyang Zhang, Ernie Tian, Zhiguo Tao, Xiaofei Han, Xiaotong Ding
ICLR 2026 Workshop on Multimodal Algorithmic Reasoning (MALGAI)
First author Accept · Poster
OpenReview ↗
2026

LLM-VR: A Generative Framework for Biometric-Aware Personalization in Extended Reality

Bingli Zhang, Xinyu Wang, Xingyu Liu, Guozhong Zhang, Yifan Wang, Yangyang Zhang, Zhang Chengbiao, Yixin Wang, Jin Cheng, Gan Shen, Zhen Tian, Zhiguo Tao
CVPR 2026 Workshop on Generative XR & Identity (GenXR-ID)
Second author · first author is the supervisor Accept · Oral
OpenReview ↗
2026

Consensus Matrix: A Role-Specialized Multi-Agent Framework for Structured Collaborative Decision-Making in Agentic Visual Media Workflows

Bingli Zhang, Xinyu Wang, Hsiang Lun Kao, Guozhong Zhang, Yijian Wu, Chenkai Gao, Yifan Wang, Zhengda, Ning Lyu, Kaijie Chen
CVPR 2026 Workshop on Agentic Audiovisual Media (AAVM)
Second author · first author is the supervisor Accept · Oral
OpenReview ↗
2026

Physics-Guided Multimodal Multi-Agent Learning for Intelligent Transportation Systems

Zhen Tian, Yaqiong Zhang, Zhihao Lin, Fujiang Yuan, Yijun Lu, Wangjie Lang, Xinyu Wang, Ning Lyu, Zhiguo Tao, Kaijie Chen, Aaron Wang
ICLR 2026 Workshop AIWILD
Co-author Published
OpenReview ↗
2026

Multi-Agent Collaborative Framework for Intelligent IT Operations: An AOI System with Context-Aware Compression and Dynamic Task Scheduling

Yixin Wang, Yingxin Su, Bingli Zhang, Zishan Bai, Guozhong Zhang, Junzhao Jiang, Xinyu Wang, Zhang Chengbiao, Yifan Wang, Xiang Luo, Jin Cheng, Ernie Tian, Xiaotong Ding
ICLR 2026 Workshop MemAgents
Co-author Published
OpenReview ↗
2026

Cognitive Digital Twin Framework: Modeling and Real-Time Decision Making

Yangyang Zhang, Mengtong Li, Xinyu Wang, Zhihao Lin, Xiang Luo, Ernie Tian, Ning Lyu, Zhiguo Tao, Xiaotong Ding, Chuanzhen Wang
ICLR 2026 Workshop on World Models
Third author Published
OpenReview ↗
Patents (Granted)

A Longitudinal Control Method for Intelligent Vehicles Based on Dynamic Feature Learning Optimization

Patent No. ZL202511097394.1 · Granted 2025
Granted

A Target Detection Method Based on LiDAR and Machine Vision Fusion

Patent No. CN 115032651 B · Granted Apr 2024
Granted
§ 04 / Projects

Research projects

Fully Automated Parking System Based on High-Accuracy Environmental Perception and High-Precision Localization

Yangtze River Delta Joint Research Programme. Trajectory planning and control strategy optimisation for fully automated parking, with system integration meeting functional-safety and AUTOSAR requirements.

Yangtze River Delta
Joint Programme

High-Accuracy Environment Recognition and Parking Trajectory Planning & Control in Complex Environments

National-level research project on perception, planning, and control for parking systems in cluttered, dynamic urban scenarios.

National Level

Heterogeneous Multi-Source Fusion and Adaptive Environmental Perception Based on Camera, LiDAR and Radar

Provincial-level research on sensor-fusion architectures and adaptive perception for autonomous driving. Outputs include cross-modal transformer fusion modules and joint calibration methods.

Provincial Level

Intelligent Monitoring and Active Warning System for Hazards in Vehicle Cabin

National College Students' Innovation and Entrepreneurship Training Program. Designed an integrated monitoring-processing-warning architecture; led sensor integration, embedded firmware, and PCB layout. Resulted in 1 publication and 1 granted patent.

MOE, China · 2017–2020
Project Leader
§ 05 / Talks

Invited talks

2023.10
Invited Speaker — Perception Fusion Algorithms for High-Level Autonomous Driving
2023 National Graduate Academic Forum on Intelligent Electric Vehicles · School of Vehicle and Mobility, Tsinghua University
2023
Speaker — National Science & Technology Programme Achievements Roadshow
Hefei–Wuhu–Bengbu High-Tech Zone Roadshow on Next-Generation Information Technologies
2023
Invited Guest — 2nd Forum on Industrial Chain & Innovation Chain Collaborative Development
Featuring keynote by Academician Keqiang Li (Tsinghua)
§ 06 / Experience

Education & positions

2025.11 — 2026.11
Visiting PhD Student
James Watt School of Engineering, University of Glasgow
2021.09 — Present
PhD in Vehicle Engineering · combined M.S.–Ph.D. programme
School of Automotive and Transportation Engineering, Hefei University of Technology
2016.09 — 2020.06
B.Eng. in Vehicle Engineering
School of Automotive and Transportation Engineering, Hefei University of Technology
§ 07 / Awards

Honours & awards

IEEE Graduate Student Member
RAS · Young Professionals · 2026
First-Class Academic Scholarship
Hefei University of Technology · 2022–2023
National Endeavor Scholarship
国家励志奖学金 · 2017–2018
Project Leader — National Innovation & Entrepreneurship Training Program
Ministry of Education, China · 2017
Class Monitor & Counselor Assistant
PhD Class of 002, HFUT
Outstanding Student Cadre
Hefei University of Technology
Merit Student (Sānhǎo)
Hefei University of Technology · multiple years
Cumulative GPA
93.58 / 100 — combined undergraduate & doctoral
§ 08 / Skills

Toolkit

Programming
Python · C · MATLAB
Simulation & Modelling
Simulink · CarSim · ROS
Languages
Chinese (native)
English (CET-6 · PETS-5/WSK)
§ 09 / Contact

Get in touch

Primary Email · HFUT
xinyuwang@mail.hfut.edu.cn
Glasgow Email
3173751W@student.gla.ac.uk
Profiles
ORCID · GitHub
Mailing Address
School of Automotive and Transportation Engineering
Hefei University of Technology
193 Tunxi Road, Baohe District
Hefei 230009, Anhui, China
Currently visiting
James Watt School of Engineering
University of Glasgow, Glasgow, Scotland, UK