George Pu

U.S. National Science Foundation, University of Florida

Papers

Total Citations

H-Index

About

George Pu is a robotics and artificial intelligence researcher whose work bridges perception, control, and situational awareness. His primary research areas include multi-robot formation control, deep reinforcement learning (DRL), and augmented reality (AR) for spatial sensing. Pu’s most impactful contribution is his 2019 paper on adaptive leader-follower formation control and obstacle avoidance using DRL, which has garnered 24 citations. In this work, he introduced a novel methodology that separates vision-based control into distinct perception and controller modules, enabling DRL agents to be trained without complex physics or 3D modeling—a significant step toward practical, scalable robot swarms. He also advanced situational awareness for search and rescue with his 2021 paper on real-time digital twin modeling of indoor spaces, which uses sensor-based mapping and AR to “see through walls” and provide enhanced visual assistance in complex built environments. With a total of 38 citations across his most-cited works, Pu is recognized for developing computationally efficient solutions that make autonomous systems more adaptable and deployable in real-world scenarios. His research holds promise for applications ranging from disaster response to collaborative robotics.