Detection of Thrust Vector Control Nozzle Failures Using Multiple Model Adaptive Estimation
Abstract
We investigated the feasibility of detecting the failure of a thrust vector control nozzle, like those proposed on advanced concepts of air-to-air missiles (Figure 1), using a modified Multiple Model Adaptive Estimation (MMAE) algorithm with a Neyman-Pearson based Hypothesis Testing Algorithm (NPHTA). These nozzles are currently being researched as a means to allow stable flight control at high angles of attack. It is likely that the missile would tumble if a nozzle failed during a high angle of attack maneuver, which could produce catastrophic results if the nozzle failure occurred as the missile is separating from the aircraft. Our goal was to determine if the NPHTA version of the MMAE could detect a nozzle failure during the separation flyout maneuver, prior to any high angle of attack maneuvers. The specific failure modes that we examined were single failures of a nozzle valve that was stuck either open or closed. We found that under idea conditions the NPHTA MMAE could detect a nozzle valve failure within one clock cycle (0.02 seconds) during this initial separation flyout and thus could test the nozzle valves prior to high angle of attack maneuvers.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1997
- Accession Number
- ADA356537
Entities
People
- Johnny H. Evers
- Peter D. Hanlon
Organizations
- Air Force Research Laboratory