Systemic Generative Engineering

Abstract

This report details research performed under this Defense Advanced Research Projects Agency Symbiotic Design for Cyber-Physical Systems effort titled "Systemic Generative Engineering." The effort focused on technical area 1, aiming to develop algorithms and frameworks enabling the creation of Artificial Intelligence (AI) co-designers. In particular, this effort sought to create a modular, open-source AI toolchain to: (i) construct, (ii) compose, and (iii) explore vast design spaces that arise in cyber-physical systems (CPS). The design space construction AI ingests previous designs and heterogeneous data, using machine learning techniques and knowledge graphs to capture explicit and implicit knowledge (e.g., mappings of requirements to functions, functions to components, components to configurations). The design space construction process accelerates the acquisition and curation of knowledge from existing design corpora; a process that today is mostly manual. The design composition AI generatively synthesizes design alternatives (conventional and novel) and assists the engineers with the automatic generation of evaluation pipelines, allowing engineers to spend their time creating, rather than tweaking tools and models. Finally, the design space exploration AI enables the inverse design of complex CPS driven by a set of target performance metrics and constraints, enabling a capability that is very difficult for experts: finding unintuitive and novel design alternatives that exhibit desired performance and behavior characteristics.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Feb 27, 2024
Accession Number
AD1222407

Entities

People

  • Arun Ramamurthy

Organizations

  • Siemens

Tags

Fields of Study

  • Computer science
  • Engineering

Readers

  • Distributed Systems and Data Platform Development
  • Software Engineering
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • Cyber
  • Space