A simulation-based approach to modeling component interactions during design of flapping wing aerial vehicles

Abstract

A new flapping wing aerial vehicle (FWAV) simulation methodology is presented that combines models of the key subsystems: (1) the actuator, (2) the battery, and (3) the wings. This approach captures component interactions that are inherently coupled in order to realize system-level designs for optimal system performance. The approach demonstrates that coupling between wing sizing, flapping motions, and loading conditions propagate into the motor–battery model to alter system-level performance properties. For the actuator subsystem model, a generalized servo motor using empirically derived coefficients to describe torque and angular velocity bandwidth in terms of voltage and current. This model is coupled with a lithium polymer battery model accounting for the nonlinear voltage drop and capacity derating effects associated with loading conditions. For aerodynamic predictions of the wing subsystem, a blade element model for predicting aerodynamic forces is coupled with an elastic wing deformation model that accounts for bending and twisting of the blade elements. System-level performance is then modeled in a design case study by coupling all of the subsystem models to account for relevant interactions, which generates a design trade space spanning a range of wing sizes, airspeeds, and flapping condition. The results from the simulation offer insight into vehicle configuration settings that provide maximum performance in terms of lift and endurance for the Robo Raven II flapping wing aerial vehicle. Experimental validation of the modeling approach shows good predictive accuracy. In addition, the presented framework offers a generalized approach for coupling interacting subsystems to improve overall predictive accuracy and identify areas where component-level improvements may offer system-level performance gains.

Surrogate Modeling of Nonlinear Dynamic Systems: A Comparative Study

“…While the sophisticated computer simulation models significantly increase the prediction accuracy, the required computational effort is also drastically increased, which poses challenges to the associated analysis, design, model updating (e.g., a digital twin), and model predictive control [9].…”

Modeling, simulation, and trade‐off analysis for multirobot, multioperator surveillance

“…The performance parameters of the FPWs in our simulation are given in Table 3. Ornithopters are still an immature research area, so we have envisioned future improvements in battery life, but the other values are realistic reflections of state‐of‐the‐art ornithopters, such as Robo Raven 47–49 …”

A Retrospective of Project Robo Raven: Developing New Capabilities for Enhancing the Performance of Flapping Wing Aerial Vehicles

“…To enhance the predictions, we made critical adjustments to enforce compatibility with the drive system, incorporated more sophisticated unsteady aerodynamic predictions, and introduced a model to compute elastic wing deformations. These modifications account for differences between stiff and flexible wings, as well as between ideal and actual wing motions, and have been validated in a similarly sized FWAV under comparable flight conditions [56].…”