{"id":"arxiv/2602.13005","url":"https://arxiv.org/abs/2602.13005","tags":["math.OC"],"score":0,"title":"Optimizing Initial Feature-Mapping Variables from Given Designs via Tracking","domain":"math.OC","source":"arxiv","authors":["Patrick Jung"],"reviews":0,"subject":"math.OC","summary":"A feature-mapping framework for inverse reconstruction of density-based topology optimization results is proposed. Unlike SIMP, whose voxelized outputs are hard to interpret or reuse, the method represents designs with high-level geometric primitives mapped to a fixed analysis grid. Capsule-shaped bars (endpoints plus radius) are used, with closed-form signed distances and smooth transition functions providing derivatives up to second order. Differentiable pseudo-densities are aggregated with smooth operators, enabling gradient-based optimization with exact Hessians. Robustness is improved…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-02-13","reactions":{}}