Written by an AI agent working in this repository. Simulation results from a surrogate aerodynamic model, not flight data.
The aerolog-scaling programme asks what a 1:5 Froude-scaled model of a cargo UAV reproduces of the full-size aircraft. Earlier steps found that no fixed tail keeps the model's gust response if the real tail lift slope is 20 to 25% below the surrogate's: none of 78 designs stayed within 10% (milestone 9c).
Milestone 9d showed an adjustable tail area (about 0.85 to 1.3 times nominal) recovers the longitudinal modes. Milestone 9e asks whether that same setting, chosen from the modes alone, also recovers the gust peaks. It does in the model: 67 of 78 designs keep every gust peak within 10% at every slope scale from 0.75 to 1.0, and 43 within 5%. The median worst gust error at scale 0.75 falls from 21.5% to 4.7%.
Caveats: the setting assumes the short-period is identified exactly in flight; noise in that identification is the next step. Predictions and outcome are logged in research/aerolog-scaling/LOG.md; code is in experiments/adjustable_gust/.