Written by an AI agent working in this repository. Numbers come from research/marine-hull/experiments/results_glider.txt (19,683 swept cases, under a second). Preliminary and simulation only.
The marine-hull programme asks which body carries a survey payload furthest per kWh. Hypothesis H4 said a glider out-ranges a same-mass, same-battery propeller AUV at 0.25 m/s but not above 1 m/s, with a crossover between 0.4 and 0.9 m/s.
What the model found. With the same hotel load for both vehicles, the glider has the longer best range in only 5.9% of cases. At five times the AUV's hotel load it wins 89%; at twenty times, all of them. A glider's pump work per horizontal metre is small, but a propeller AUV at its best speed is already cheap, so the gap is a hotel-load gap. The pump work, B tan(gamma)/(2 eta), does not depend on dive depth. The predicted crossover was refuted: it falls in 0.4 to 0.9 m/s in 0.4% of cases, because a 10 N pump caps glider horizontal speed at 0.13 to 0.70 m/s (median 0.30), so the "crossover" mostly measures a speed ceiling.
What it does not show. No glider coefficient is sourced. The Seaglider and glider-flight papers were unreachable through the sandbox proxy, so every coefficient was swept and the hotel multiplier was added after seeing the first result. H4 stays open until measured power budgets are found.