AirCopBench: A Benchmark for Multi-drone Collaborative Embodied Perception and Reasoning
Evaluating collaborative perception and reasoning in multi-drone settings, including degraded visual conditions.
Hi, I’m Yingfeng Chen, a third-year undergraduate at Weiyang College, Tsinghua University. I began my studies in 2024, studying Mathematics and Physics + Interdisciplinary Engineering.
I am interested in embodied AI—how intelligent agents perceive, understand, and interact with the physical world. In particular, I am curious about how embodied agents navigate complex urban environments, integrate observations into an understanding of their surroundings, and make decisions as the world changes.
I am also interested in how objects and their relationships can be represented, how robots develop spatial and semantic understanding, and how perception, language, and action can be connected. More broadly, I hope to explore how robots learn through interaction and use that experience to reason, plan, and adapt to unfamiliar situations.
Evaluating collaborative perception and reasoning in multi-drone settings, including degraded visual conditions.
Answering open-vocabulary questions through active urban exploration with a hierarchical Planner–Manager–Actor architecture.
Psychology-driven language-model agents for interpretable modeling of panic emotions during disaster events.