Satoru Kato

National Institute of Technology, Matsue College

Papers

Total Citations

H-Index

About

Satoru Kato is a pioneering researcher in the intersection of deep reinforcement learning and autonomous robotics, with a primary focus on enabling mobile robots to learn complex behaviors through vision-based systems. His landmark 2017 study, “A study on vision-based mobile robot learning by deep Q-network,” which has accumulated 78 citations, stands as a foundational contribution to the field. In this work, Kato demonstrated how Deep Q-Networks (DQN)—a method that approximates action-value functions using Convolutional Neural Networks and updates them via Q-learning—could be effectively applied to robot behavior acquisition in simulation environments. By constructing a virtual setting where a mobile robot learned to navigate and interact using only visual input, Kato provided a proof-of-concept that deep reinforcement learning could bridge the gap between raw sensor data and intelligent action. His subsequent experimental study, though less cited, reinforced these findings by validating the approach’s robustness. Kato’s work has been instrumental in advancing the practical deployment of DQN in robotics, inspiring further research into end-to-end learning for autonomous systems. His contributions remain a key reference for students and researchers exploring deep RL for real-world robotic applications.