Sang-hyeok

Papers

Total Citations

H-Index

About

Sang-hyeok is a robotics researcher whose work focuses on advancing path planning and decision-making algorithms for autonomous mobile robots. His most notable contribution is the development of the extended Dyna-Q algorithm, which addresses a critical limitation of the standard Dyna-Q approach: the inefficient, blind exploration that occurs during initial learning episodes. By incorporating a maximum likelihood model of all state-action pairs, Sang-hyeok’s method significantly improves learning efficiency, enabling robots to navigate more intelligently from the very first steps. This work, published in 2011 and cited 10 times, has provided a practical foundation for reinforcement learning in real-world robotic navigation. Sang-hyeok’s research sits at the intersection of machine learning and robotics, demonstrating how model-based enhancements can make reinforcement learning algorithms more robust and applicable to dynamic environments. His contributions are particularly valuable for students and researchers interested in efficient autonomous systems, offering a clear example of how theoretical improvements in algorithm design translate directly into better robot performance.