Tuning of a rigid-body dynamics model of a flapping wing structure with compliant joints

Abstract

A method for validating rigid-body models of compliant mechanisms under dynamic loading conditions using motion tracking cameras and genetic algorithms is presented. The compliant mechanisms are modeled using rigid-body mechanics as compliant joints (CJ): spherical joints with distributed mass and three-axis torsional spring dampers. This allows compliant mechanisms to be modeled using computationally efficient rigid-body dynamics methods, thereby allowing a model to determine the desired stiffness and location characteristics of compliant mechanisms spatially distributed into a structure. An experiment was performed to validate a previously developed numerical dynamics model with the goal of tuning unknown model parameters to match the flapping kinematics of the leading edge spar of an ornithopter with contact-aided compliant mechanisms (CCMs), compliant mechanisms that feature self-contact to produce nonlinear stiffness, inserted. A system of computer motion tracking cameras was used to record the kinematics of reflective tape and markers placed along the leading edge spar with and without CCMs inserted. A genetic algorithm was used to minimize the error between the model and experimental marker kinematics. The model was able to match the kinematics of all markers along the spars with a root-mean-square error (RMSE) of less than 2% of the half wingspan over the flapping cycle. Additionally, the model was able to capture the deflection amplitude and harmonics of the CCMs with a RMSE of less than 2 deg over the flapping cycle.