Baris Serhan

University of Lincoln, University of Nottingham, University of Manchester

Papers

Total Citations

H-Index

About

Baris Serhan is a pioneering researcher at the intersection of neuromorphic computing, robotic manipulation, and human-robot interaction. His work is anchored in three key areas: developing brain-inspired spiking neural networks (SNNs) for robotic perception, advancing non-prehensile manipulation for object segmentation, and integrating Theory of Mind (ToM) into trust-aware robot policies. Serhan’s most-cited paper (28 citations) introduces an on-chip SNN implemented on Intel’s Loihi neuromorphic chip, enabling the iCub humanoid robot to estimate its head pose and represent scenes with remarkable energy efficiency—a landmark achievement in deploying SNNs for real-time robotic control. His subsequent work on “Push-to-See” (14 citations) leverages deep Q-learning to teach robots to physically nudge cluttered, textureless objects, dramatically improving instance segmentation when traditional vision fails. Most recently, his “ToP-ToM” framework (7 citations) pioneers a cognitive architecture that allows robots to infer human beliefs and intentions, adjusting their actions to build trust and facilitate seamless collaboration. Through these contributions, Serhan is shaping a future where robots perceive, manipulate, and interact with humans in more intelligent, adaptive, and socially aware ways.