C. James Taylor

Papers

Total Citations

H-Index

About

C. James Taylor is a researcher whose work spans the fields of robotics, control systems, and machine learning, with a particular focus on dynamics modeling for complex robotic systems. His research addresses fundamental challenges in model-based control, specifically developing methods to accurately estimate forward and inverse dynamics in sophisticated robotic platforms — including hydraulically-driven robots, artificial muscle actuators, and systems navigating varied contact situations. A central contribution of his work is the development of Action-Conditional Recurrent Kalman Networks, an innovative approach that combines recurrent neural network architectures with Kalman filtering principles to learn dynamics models in settings where traditional analytic models are unavailable or insufficiently accurate. This hybrid methodology bridges classical control theory with modern deep learning, offering a principled framework for robots operating under real-world uncertainty. While his work is in its early citation stages — reflecting relatively recent publication — his research addresses pressing and practically significant problems in robotics that are attracting growing attention from the community. Students and researchers working at the intersection of machine learning and robot control will find his contributions particularly relevant to challenges in data-driven modeling, adaptive control, and autonomous robotic manipulation.