No description available.

Regime recognition is an important tool used in creation of usage spectra and fatigue loads analysis. While a variety of regime recognition algorithms have been developed and deployed to date, verification and validation (V&V) of such algorithms is still a labor intensive process that is largely subjective. The current V&V process for regime recognition codes involves a comparison of scripted flight test data to regime recognition algorithm outputs. This is problematic because scripted flight test data is expensive to obtain, may not accurately match the maneuver script, and is often used to train the regime recognition algorithms and thus is not appropriate for V&V purposes. In this paper, a simulation-based virtual pilot algorithm is proposed as an alternative to physical testing for generating V&V flight test data. A “virtual pilot” is an algorithm that replicates a human’s piloting and guidance role in simulation by translating high level maneuver instructions into parameterized control laws. Each maneuver regime is associated with a feedback control law, and a control architecture is defined which provides for seamless transitions between maneuvers and allows for execution of an arbitrary maneuver script in simulation. The proposed algorithm does not require training data, iterative learning, or optimization, but rather utilizes a tuned model and feedback control laws defined for each maneuver. As a result, synthetic HUMS data may be generated and used in a highly automated regime recognition V&V process. In this thesis, the virtual pilot algorithm is formulated and the component feedback control laws and maneuver transition schemes are defined. Example synthetic HUMS data is generated using a simulation model of the SH-60B, and virtual pilot fidelity is demonstrated through both conformance to the ADS-33 standards for selected Mission Task Elements and comparison to actual HUMS data.