Hello! I am a Research Associate and Ph.D. candidate at the Professorship of Autonomous Vehicle Systems, School of Engineering and Design, Technical University of Munich, Germany, under the supervision of Professor Dr.-Ing. Johannes Betz. My research focuses on foundation-model-based scenario generation and analysis for autonomous driving. I have recently graduated with the double Master degree in Mechatronics and Robotics with high distinction, as well as Development, Production, and Management in Mechanical Engineering with distinction from the same university TUM in 2023.
Read moreShow less
During my master's studies, I conducted my research at the Chair of Information-oriented Control under the supervision of Dr. Junjie Jiao and Professor Dr.-Ing. Sandra Hirche, where I completed my master's thesis. Prior to that, I worked as a student research assistant and wrote my term's thesis at the Chair of Automotive Technology under the supervision of Professor Dr.-Ing. Markus Lienkamp.
I earned my Bachelor's degree in Mechanical Design, Manufacture, and Automation with Excellent Graduate honors at School of Mechanical Engineering, Hefei University of Technology, China in 2017.
News
- 2026-06: Our paper EgoDyn-Bench: Evaluating Ego-Motion Understanding in Vision-Centric Foundation Models for Autonomous Driving was accepted at ECCV 2026. [Project Page] (Co-author) [New!]
- 2026-06: Our paper Target-Bench: Can Video World Models Achieve Mapless Path Planning with Semantic Targets? was accepted at ECCV 2026. (Co-author) [New!]
- 2026-06: Our paper Chat2Scenic: An Iterative RAG-Based Framework for Scenario Generation in Autonomous Driving was accepted at IEEE IROS 2026. (First author) [New!]
- 2026-06: Attended IEEE ICRA 2026 in Vienna, Austria.
- 2026-05: Our paper Can VLMs Unlock Semantic Anomaly Detection? A Framework for Structured Reasoning was accepted at IEEE ITSC 2026. (Co-author)
- 2025-12: Attended the 2025 German Autonomy Summit (GAS) in Munich, Germany.
- 2025-11: Visited the Australian Centre for Robotics at the University of Sydney with my colleague Marc Kaufeld.
- 2025-11: Attended IEEE ITSC 2025 in Gold Coast, Australia, and gave presentations at the LLVM-AD Workshop and the Interaction-driven Behavior Prediction and Planning for Autonomous Vehicles Workshop.
- 2025-09: Our paper NuRisk: A Visual Question Answering Dataset for Agent-Level Risk Assessment in Autonomous Driving was accepted at IEEE ICRA 2026. (First author)
- 2025-06: Our survey paper Foundation Models in Autonomous Driving: A Survey on Scenario Generation and Scenario Analysis was accepted by the IEEE Open Journal of Intelligent Transportation Systems. (First author)
- 2025-04: Our paper From Words to Collisions: LLM-Guided Evaluation and Adversarial Generation of Safety-Critical Driving Scenarios was accepted at IEEE ITSC 2025. (First author)
- 2024-11: Attended the 2024 Conference on Robot Learning (CoRL) in Munich, Germany.
- 2024-10: Joined the Professorship of Autonomous Vehicle Systems (AVS) at TUM as a doctoral candidate.
Publications — Highlights
- An iterative, RAG-based framework that turns natural-language descriptions into executable driving scenarios using retrieval-augmented LLMs.
- A benchmark for evaluating ego-motion understanding in vision-centric foundation models for autonomous driving.
- A benchmark probing whether video world models can perform mapless path planning toward semantic targets.
- A comprehensive survey of foundation models, LLMs, VLMs, MLLMs, Diffusion and world models, for scenario generation and scenario analysis in autonomous driving.
- A VQA dataset and benchmark for agent-level risk assessment in autonomous driving.
- LLM-guided evaluation and adversarial generation of safety-critical driving scenarios.
- Distributed control design for H2 suboptimal containment of multi-agent systems.
- H2 suboptimal leader–follower consensus control for multi-agent systems.