My research interests primarily lie in Robotics, Reinforcement Learning, Computer Vision and Control Theory. My current research focuses on addressing fundamental challenges in grounding large language models (LLMs) in robotics, with the goal of enabling intelligent agents to effectively perceive, reason about, and interact with the physical world.
π₯ News
[July 2026] Our paper about adaptive uncertainty quantification is accepted by CDC 2026.
[April 2026] Our paper about safe model-based RL is accepted by IROS 2026.
[April 2026] Our paper about Time-aggregated Connectivity is accepted by RSS 2026.
[April 2026] Our paper about Adaptive PT-CBF is accepted by IFAC 2026.
[Jan. 2026] Our paper about geometry-aware CBF is accepted by ICRA 2026.
[Jun. 2025] I will present our Multi-robot Deadlock Resolution paper in MMLS 2025.
[Jan. 2025] Our paper about Multi-robot Deadlock Resolution is accepted to ICRA' 25.
[Jan. 2025] Our paper about Sample-efficient Safe Reinforcement Learning is accepted to ICRA' 25.
We present a novel decentralized framework that ensures both efficient task execution and deadlock
resolution for multi-agent systems, which enables co-optimization
between the task-related controller and deadlock resolution
controller, yielding smoother robotsβ motion with improved overall task execution efficiency
**Courteous MPC for Autonomous Driving with CBF-inspired Risk Assessment**
Yanze Zhang, Yiwei Lyu, Sude E. Demir, Xingyu Zhou, Yupeng Yang, Junming Wang, Wenhao Luo
IEEE International Conference on Intelligent Transportation Systems (ITSC), 2024
we develop an extension of the CBF-inspired risk evaluation framework that takes into account both noisy observed positions and motions, which are then integrated with MPC to ensure smooth and robust decision-making for AVs.
All Publications
Adaptive Deadlock Avoidance for Decentralized Multi-Agent Systems via CBF-inspired Risk Measurement.
Yanze Zhang, Yiwei Lyu, Siwon Jo, Yupeng Yang, and Wenhao Luo
Accepted to the 2025 IEEE International Conference on Robotics and Automation (ICRA' 25), 2025