Mechanical Neural-Network Architecture Materials that Learn

Abstract

The objective of this research is to apply the concept of artificial neural networks to enable the creation of a new kind of architectured material called mechanical neural-network (MNN) architectured materials that can learn desired properties via a complex web of active flexible elements (AFEs) that constitute the materials microstructure. Although significant research has been conducted toward enabling advanced materials that utilize active elements to achieve programmable properties, many scenarios exist where a materials environment may change and it is necessary that the material can autonomously adapt its properties accordingly to successfully fulfill its desired purpose. In such scenarios, designers rarely have the time o rknowledge of each environmental change to program and upload new control instructions to change the properties as required. Thus, it is necessary that the material can learn to change its properties on its own.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 10, 2022
Accession Number
AD1230348

Entities

People

  • Jonathan B. Hopkins

Organizations

  • University of California, Los Angeles

Tags

Readers

  • Nanocomposite Materials Science
  • Neural Network Machine Learning.
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks