Updating Rotorcraft Simulation Environments by Using “Black-Box” Input Filters

Abstract

Rotorcraft simulation environments such as training simulators or engineering simulations require high-fidelity models to be suitable for tasks such as pilot training, handling qualities analysis, and flight control system development. Deficits in model fidelity can arise from inaccurate data and unmodeled higher-order dynamic effects. Mitigating these deficits through physics-based modeling usually requires high effort. Therefore, a method is presented that improves a baseline simulation model by a “black-box” (i.e. non-physical) low-order input filter. The baseline can be a linear model or a nonlinear simulation model. Several options for deriving the input filter are elaborated and demonstrated in this paper using data from the CH-47, AH 135, and Bell 412 helicopters. In all cases the simulation fidelity was improved by approximately a factor of two.