Xiyang Wu

Georgia Institute of Technology, University of Maryland, College Park

Papers

Total Citations

H-Index

About

Xiyang Wu is a robotics and artificial intelligence researcher whose work spans multi-agent systems, robot navigation, and the integration of large language models into autonomous systems. He is perhaps best known for developing **FireCommander**, an interactive, probabilistic multi-agent simulation environment designed to support joint perception-action research in heterogeneous robot teams. This platform, which has garnered over 10 citations across two related publications, enables researchers to study cooperative robotics in complex, dynamic firefighting scenarios — providing a valuable benchmark for the broader multi-agent research community. Wu's more recent contributions reflect a growing interest in language-guided robotics. His work on LANCAR explores how large language models can provide contextual awareness to help robots navigate unstructured terrains, addressing one of autonomous mobility's most persistent challenges. Complementing this, his investigations into the vulnerabilities of LLM- and vision-language-model-controlled robotic systems raise critical safety questions, demonstrating that impressive task performance can mask dangerous fragility under slight input variations. Together, Wu's research positions him at a compelling intersection of trustworthy AI, embodied intelligence, and human-robot collaboration — areas of rapidly growing importance as autonomous systems move into real-world deployment.