A new book titled Applications of Neural Networks in High Assurance Systems has been edited by Dr. Johann Schumann (SGT) and Yan Liu (Motorola Labs), and contains contributions from other Division staff. This book presents relevant theoretical research and recent advances in industrial application and certification of neural network-based learning systems in safety-related areas like damage adaptive aircraft control, fault detection in automotive engines, and control of submarines, fuel cells, and oil-blending. The book is published by Springer, includes a preface by P. Werbos, and is available at Springer, Amazon, Borders, Barnes & Noble, and other outlets.
BACKGROUND: Applications of Neural Networks in High Assurance Systems addresses a key field of neural network technology: the methods employed to pass rigorous verification and validation (V&V) standards required for many safety-critical applications. The book looks at the types of evaluation methods developed across many fields and how to pass V&V, as well as a new adaptive structure of V&V that is developed, which is different than both the six sigma methods generally used for large-scale systems and the theorem-based approach used for simplified component subsystems.
CONTRIBUTORS: A. Annaswami (MIT), J. Barhorst (Boeing), B. Cukic (WVU), D. Djurdjanovic (UTAustin), E. Fuller (WVU), M. Gheorghiu (NC A&T State U),
P. Gupta (UARC), S. Jacklin (NASA Ames), J. Jang (MIT), E. Lavretsky (Boeing), X. Li (Mexico DF), J. Liu (GM Research), Y. Liu (Motorola), K. Marko (ETAS Inc),
A, Mehrabian (Concordia U), M. Menhaj (Amirkabir U), N. Nguyen (NASA Ames), J. Ni (U Michigan), A. Rezazadeh (Shahid Beheshti U), J. Schumann (SGT),
M. Sedighizadeh (Shahid Beheshti U), T. Smith (Boeing), D. Song (Beijing Jiaotong U), J. Urnes (Boeing), L. Weng (NC A&T State U), W. Yu (Mexico DF), X-H. Yu (CalPoly), S. Yerramalla (Pratt&Whitney)
PROGRAM FUNDING: Editing work has been funded by the NASA Aeronautics ARC-AF and NASA Space Grant.
Contact: Johann Schumann