OmniAD: Data-driven Omni-directional Aerodynamics

摘要

This paper introduces “OmniAD,” a novel data-driven pipeline tornmodel and acquire the aerodynamics of three-dimensional rigidrnobjects. Traditionally, aerodynamics are examined through elaboraternwind tunnel experiments or expensive fluid dynamics computations,rnand are only measured for a small number of discreternwind directions. OmniAD allows the evaluation of aerodynamicrnforces, such as drag and lift, for any incoming wind direction usingrna novel representation based on spherical harmonics. Our datadrivenrntechnique acquires the aerodynamic properties of an objectrnsimply by capturing its falling motion using a single camera. Oncernmodel parameters are estimated, OmniAD enables realistic realtimernsimulation of rigid bodies, such as the tumbling and glidingrnof leaves, without simulating the surrounding air. In addition, wernpropose an intuitive user interface based on OmniAD to interactivelyrndesign three-dimensional kites that actually fly. Various nontraditionalrnkites were designed to demonstrate the physical validityrnof our model.