Validate transfer to independent regolith realizations and continuous modulus values

Validate whether the early penetrator-response calibration transfers across independent granular-bed realizations and supports interpolation to continuous, previously unsampled particle Young’s-modulus values.

Background

The Young’s-modulus ensemble uses 17 penetration locations within a single parent-bed realization. Leave-one-location-out validation therefore measures transfer to an unseen coordinate within that same bed, rather than generalization across independently generated beds. The analysis also classifies only the five sampled modulus levels or their broad groups; continuous interpolation between unsampled modulus values is not tested.

References

Because neighboring coordinates can share particles and surrounding structure, these metrics characterize transfer to an unseen coordinate within the present bed; they are not independent-bed generalization accuracies. The class-grouped results support coarse narrowing of the admissible Young's-modulus range under the present model conditions, whereas interpolation to unseen continuous $E$ values remains untested.

— Constraining Particle Stiffness in Asteroid Regolith from Early-Time High-Speed Penetrator Dynamics under Local Granular Variability  (2610.01470 - Wang et al., 1 Oct 2026) in Section 3.2, “Young’s-modulus response under local configurational variability”