Abstract:In biological systems, sensing is not performed by the brain alone: the body deforms, vibrates, and filters external stimuli before they are transduced into neural signals. In engineered systems, this processing burden is placed largely on electronics and computation, while the mechanical body is usually designed only for strength and stability. Here, we present sensing intelligence as a trainable property of the body. We show that the geometry of a metamaterial can be optimized to reshape external stimuli into internal signals that are easier for a neural network to interpret. Rather than hand-designing this physical preprocessing, we let the neural network train its own body for sensing by backpropagating the sensing loss to the body's design parameters through differentiable simulation. Across numerical and experimental sensing scenarios, the optimized body improves sensing accuracy by up to fivefold or reduces the number of required electronic sensors by nearly an order of magnitude.
| Subjects: | Neurons and Cognition (q-bio.NC); Materials Science (cond-mat.mtrl-sci); Artificial Intelligence (cs.AI); Applied Physics (physics.app-ph) |
| Cite as: | arXiv:2605.23967 [q-bio.NC] |
| (or arXiv:2605.23967v1 [q-bio.NC] for this version) | |
| https://doi.org/10.48550/arXiv.2605.23967 arXiv-issued DOI via DataCite |
From: Bolei Deng [view email]
[v1]
Wed, 13 May 2026 01:55:34 UTC (27,661 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。