YF22 Model With On-Board On-Line Learning Microprocessors-Based Neural Algorithms for Autopilot and Fault-Tolerant Flight Control Systems.

摘要

This project focused on investigating the potential of on-line learning 'hardware-based' neural approximators and controllers to provide fault tolerance capabilities following sensor and actuator failures. Following a phase of simulation studies a set of selected architectures for neural estimators and neural controllers were flown on a semi-scale YF-22 aircraft model. The YF-22 model was designed, built, and flown at research facilities at West Virginia University. Additionally, a customized electronic payload featuring these fault tolerant schemes was designed, built, tested and interfaced with the YF-22 flight control system. A series of 33 flight tests were conducted with the aircraft; the flight data confirmed the potential of neural estimators and controllers for fault tolerance purposes. Another research objective was to start addressing system requirements leading to the problem of software validation and verification for this new class of algorithms for fault tolerant flight control systems.