๐Ÿ‘จโ€๐ŸŽ“ About Me

I am a third-year Ph.D. student in the School of Vehicle and Mobility at Tsinghua University, working under the supervision of Prof. Sifa Zheng.

My research centers on traffic simulation and motion planning for autonomous driving, with a particular emphasis on creating realistic, interactive, and controllable traffic scenarios.ย I am passionate about developing reliable closed-loop simulation frameworks to accelerate the safe, robust, and scalable advancement of autonomous driving systems.

๐ŸŽ“ Educations

Tsinghua University logo
THU (Tsinghua University), Beijing, China
Ph.D. candidate in Mechanical Engineering, School of Vehicle and Mobility.
2023.09 - present
NUAA logo
NUAA (Nanjing University of Aeronautics and Astronautics), Nanjing, China
B.Eng. in Vehicle Engineering.
2019.09 - 2023.06

๐Ÿ”ฅ News

  • 2025.05: ๐Ÿ”ฅ Released RIFT! Check it out!

  • 2024.09: ๐ŸŽ‰ FREA accepted by CoRL 2024 for Oral (4.3%).
  • 2023.03: ๐ŸŽ‰ IGT accepted by Knowledge-Based Systems 2023.

๐Ÿ“ Publications

๐Ÿ”† Select Publications

arXiv 2025
sym

RIFT: Group-Relative RL Fine-Tuning for Realistic and Controllable Traffic Simulation

Keyu Chen,Wenchao Sun, Hao Cheng, Sifa Zheng

arXiv 2025

Custom badge Custom badge Custom badge Stars

RIFT achieve realistic and controllable traffic simulation by combining IL pre-training in a data-driven simulator for realism with RL fine-tuning in a physics-based simulator for controllability.

CoRL 2024
sym

FREA: Feasibility-Guided Generation of Safety-Critical Scenarios with Reasonable Adversariality

Keyu Chen, Yuheng Lei, Hao Cheng, Haoran Wu, Wenchao Sun, Sifa Zheng

Conference on Robot Learning (CoRL Oral 4.3%), 2024

Custom badge Custom badge Custom badge Stars

FREA incorporates feasibility as guidance to generate adversarial yet AV-feasible, safety-critical scenarios for autonomous driving.

KBS 2023
sym

IGT: Illumination-guided RGB-T object detection with transformers

Keyu Chen, Jinqiang Liu, Han Zhang

Knowledge-Based Systems (KBS) (SCI Q1, IF=7.4), 2023

Custom badge

IGT uses illumination intensity to guide the fusion process of multi-modality features, enabling the comprehensive utilization of cross-modality complementary information in object detection.

๐Ÿ“š All Publications

๐Ÿš€ Project

Github
sym

Highway-Adversarial

Custom badge Stars

We manipulate critical background vehicles in the Highway-Env to exhibit adversarial behaviors that challenge the ego vehicle, creating safety-critical scenarios.

๐Ÿง‘โ€๐Ÿ’ป Internships

Li Auto logo
Li Auto, Beijing, China
Closed-loop RL fine-tuning for autonomous driving.
2025.11 - present
Tsari logo
Tsari, Suzhou, China
Safety-critical scenario generation for autonomous driving.
2023.07 - 2023.08

๐Ÿ† Honors and Awards

2023.06
Outstanding Graduate (Top 1%)
Nanjing University of Aeronautics and Astronautics (NUAA).
2023.06
Outstanding Bachelor Thesis Award (Top 5%)
Nanjing University of Aeronautics and Astronautics (NUAA).
2021.12
National Scholarship (Top 1%)
Nanjing University of Aeronautics and Astronautics (NUAA).

๐Ÿ—บ๏ธ Visitor Map