Nathan Glaser

Papers

Total Citations

166

H-Index

About

Nathan Glaser is a leading researcher in collaborative perception and multi-agent robotic systems, with a focus on enabling robots to intelligently share and combine sensory data for improved environmental understanding. His most influential work, "Who2com: Collaborative Perception via Learnable Handshake Communication" (2020), has garnered over 164 citations, establishing him as a pioneer in the field. In this seminal paper, Glaser introduced a novel framework where robots learn to selectively communicate with neighboring agents through a "handshake" mechanism, dynamically deciding which observations to share to maximize collective perception accuracy. This work fundamentally advanced the problem of collaborative perception, moving beyond traditional robotics and multi-agent reinforcement learning approaches by making the communication protocol itself learnable. Glaser's contributions have significant implications for autonomous driving, drone swarms, and distributed sensing networks, where efficient and intelligent data sharing is critical. His research continues to shape how multi-robot systems can achieve superhuman perception capabilities through coordinated, selective communication, making him a key figure in the next generation of autonomous systems.