 paper airplanes have been long considered a child's plafing, but they actually contain a surprising amount of complexity. A new robotic system has been developed which can fabricate, test, analyze, and model the flight behavior of paper airplanes. It has identified three distinct types of trajectories, nose dive, glide, and recovery glide, and used them to create a Gaussian mixture model, GMM, to extract the clusters of the design space that map to each type of trajectory. This allows for both the forward and reverse design problem to be solved, as well as efficient optimization of the geometry for a given target flight distance. This article was authored by Nana Obayoshi, Kai Young, Stefan Illich, and others.