Multiple Model Adaptive Estimation Applied to the VISTA F-16 with Actuator andSensor Failures. Volume 2

摘要

A Multiple Model Adaptive Estimation (MMAE) algorithm is applied to the VariableStability In-flight Simulator Test Aircraft (VISTA) F-16 at a low dynamic pressure flight condition (0.4 Mach at 20000 ft). A complete F-16 flight control system is modeled containing the longitudinal and lateral-directional axes. Single and dual actuator and sensor failures are simulated including: complete actuator failures, partial actuator failures, complete sensor failures, increased sensor noise, sensor biases, dual complete actuator failures, dual complete sensor failures, and combinations of actuator and sensor failures. Failure scenarios are examined in both maneuvering and straight and level flight conditions. The system performance is characterized when excited by purposeful commands and dither signals. Single scalar residual monitoring techniques are evaluated with suggestions for improved performance. A Kahnan filter is designed for each hypothesized failure condition. In this thesis, thirteen elemental Kalman filters are designed encompassing: a no failure filter, left stabilator failure filter, a right stabilator failure filter, a left flaperon failure filter, a right flaperon failure filter, a rudder failure filter, a velocity sensor failure filter, an angle of attack sensor failure filter, a pitch rate sensor failure filter, a normal acceleration sensor failure filter, a roll rate sensor failure filter, a yaw rate sensor failure filter, and a lateral acceleration sensor failure filter.