Masashi Inoue

Yokohama National University, Mitsubishi Electric (Japan)

Papers

Total Citations

H-Index

About

Masashi Inoue is a pioneering researcher in the intersection of reinforcement learning and robotics, with a particular focus on how autonomous agents discover unexpected and efficient motion forms. His work centers on the application of Q-learning algorithms to mobile and space robotics, demonstrating that reinforcement learning can yield novel solutions beyond traditional programming constraints. In his most cited study (2006, 16 citations), Inoue showed how a two-dimensional mobile robot could acquire surprising, non-intuitive motion strategies through trial-and-error learning, highlighting the creative potential of AI-driven design. He further explored this phenomenon in a 2006 paper (10 citations) analyzing the learning process of a caterpillar robot, revealing how simple two-actuator systems can develop complex looping and advance actions. Earlier in his career, Inoue contributed to space robotics with a simulation system for a six-axis servo-controlled space robot (1989, 11 citations). His research provides foundational insights into emergent robot behavior, bridging reinforcement learning theory with practical robotic locomotion. Inoue’s work continues to inspire researchers in adaptive robotics and autonomous systems, demonstrating that machine learning can unlock motion forms that human engineers might never conceive.