E.B. Kosmatopoulos
Technical University of Crete, University of Southern California
Papers
2
Total Citations
49
H-Index
2
About
E.B. Kosmatopoulos is a leading figure in nonlinear system identification and intelligent control, with a career marked by pioneering work in neural network architectures for complex dynamical systems. His most influential contribution, the 2005 paper on "Identification of nonlinear systems using new dynamic neural network structures," has garnered 41 citations and remains a cornerstone in the field. In this work, Kosmatopoulos rigorously established the stability and convergence properties of recurrent high-order neural networks (RHONNs), introducing dynamical neurons that enable accurate modeling of nonlinear systems—a breakthrough that has shaped modern adaptive control and system identification methodologies. Earlier, his 1997 study on "High-order neural networks for the learning of robot contact surface shape" (8 citations) demonstrated a novel application of extended Kalman filters to estimate unknown surface parameters during robotic contact, advancing autonomous manipulation. Kosmatopoulos’s research seamlessly bridges theoretical rigor and practical robotics, with his RHONN framework influencing applications from aerospace to industrial automation. His work is essential reading for students and researchers exploring neural-network-based control, offering foundational insights into how dynamic neural structures can tackle real-world nonlinear challenges.
Research Focus
Key Achievements
Top Papers
- 1
- 2