Shaw-Ji Shiah

National Yang Ming Chiao Tung University

Papers

Total Citations

H-Index

About

Shaw-Ji Shiah is a researcher whose work centers on advancing robot learning control, particularly for multijoint manipulators. His major contributions address a fundamental challenge in robotics: the vast learning space required for general motions. In his 1997 paper (14 citations), he proposed an approach to enlarge learning space coverage, enabling controllers to handle a wider range of trajectories without repeated retraining. This work aimed to move beyond conventional subordinate controllers by making learning more autonomous and efficient. His 2002 study (2 citations) further refined this concept by introducing a novel scheme for governing similar robot motions, reducing the need to restart the learning process for each new trajectory. While his citation counts are modest, Shiah’s focus on scalability and generalization in robot learning represents an early effort to tackle persistent issues in adaptive control. His research is notable for its forward-looking emphasis on creating learning systems that can transfer knowledge across tasks—a precursor to modern meta-learning and transfer learning in robotics. For students and researchers, Shiah’s work offers a foundational perspective on the challenges of building truly adaptive robotic systems.