ADAPTIVE EXPERIMENTAL OPTIMISATION FOR SAMPLE-EFFICIENT ARMOUR DESIGN

Abstract

We present a Bayesian optimisation methodology intended to support a human expert in the design of armour systems for which limited prior knowledge/data exists and within a limiting, pre-defined experimental budget. We apply the methodology to design an armour configuration consisting of multiple plates, with multiple materials, at varying orientations and spacing, for protection against 12.7 mm APM2 and 20 mm FSP threats. The full-factorial design matrix for the defined solution space exceeds 17,500 possible solutions. With an objective to minimise system weight, we identify a solution within 102 ballistic tests (44 design iterations) that provides a weight reduction of 11.4% over expert-designed reference configurations and a mass efficiency of 1.5 relative to a monolithic RHA Class 1. The value of the demonstrated methodology is expected to increase with increasing armour (or threat) complexity.