Kaoru Kawabata

Papers

Total Citations

H-Index

About

Kaoru Kawabata is a pioneering researcher in the field of autonomous control systems for large-scale industrial machinery, with a particular focus on waste incineration plants. His work addresses the critical challenge of optimizing control policies for massive mechanical systems, such as waste cranes, that operate in complex, weakly observable environments. Kawabata’s most influential paper, "Bayesian Policy Optimization for Waste Crane With Garbage Inhomogeneity" (2020, 10 citations), introduces a groundbreaking framework that enables autonomous trial-and-error learning for slow-moving industrial equipment, significantly improving efficiency despite the unpredictable nature of waste composition. Building on this, his 2022 study on "Gaussian Process Self-triggered Policy Search in Weakly Observable Environments" (2 citations) tackles the practical limitations of sensor-poor settings, offering a robust solution for environments where state observations are minimal due to technical or cost constraints. Kawabata’s contributions are vital for advancing automation in heavy industries, demonstrating how machine learning can be adapted to real-world, physically constrained systems. His work not only enhances operational performance but also reduces human intervention in hazardous or inaccessible industrial sites, marking him as a key innovator in applied robotics and control theory.