The Department of Defense (DoD) is seeking innovative solutions through the Small Business Innovation Research (SBIR) program to develop an AI-driven system for converting 2D engineering drawings into accurate, manufacturable 3D models. This initiative aims to enhance the sustainment and operational readiness of legacy systems by addressing the challenges posed by outdated schematics, ultimately revitalizing the Defense Industrial Base (DIB) and improving mission readiness. The project will unfold in three phases, starting with feasibility assessments and culminating in a fully operational system that integrates advanced features such as digital twin creation for predictive maintenance. Interested parties should note that the solicitation is set to open on May 28, 2025, with applications due by June 25, 2025, and can find further details at the official DoD SBIR/STTR website.

## Description

This topic seeks to revolutionize the Department of Defense’s (DoD) ability to sustain legacy systems by advancing AI-driven technology for converting 2D engineering drawings into accurate, manufacturable 3D models. Many of the DoD’s aging platforms rely on outdated, often incomplete 2D schematics that are difficult for modern manufacturers to interpret. This challenge has contributed to sustainment delays, inefficiencies, and a diminishing Defense Industrial Base (DIB) as legacy parts suppliers go out of business or struggle to meet requirements. By addressing these issues, this project aims to enhance readiness, extend the operational life of aging systems, and strengthen the DIB through innovative digital engineering solutions.&nbsp;At the heart of this effort is the development of an AI-powered system capable of interpreting complex 2D schematics and creating precise 3D models suitable for manufacturing and digital twin applications. Existing AI tools are primarily optimized for artistic or generalized 3D generation and lack the precision and adaptability needed for engineering applications. This topic explores the creation of an advanced AI system that bridges this gap, enabling rapid and accurate 3D modeling for legacy systems.&nbsp;A critical component of this initiative is its integration with the existing systems. Leveraging existing secure data-sharing capabilities, this project will facilitate the development of a protected repository of engineering data, combining existing 3D CAD/CAM models with newly converted models. This centralized resource will foster collaboration among DIB partners, enhance data integrity, and ensure compliance with stringent DoD security protocols. By incorporating blockchain technology, the project will enable traceability and protect sensitive data, ensuring its use for authorized applications only.&nbsp;This initiative also addresses the growing importance of digital twins for predictive maintenance and lifecycle management of DoD systems. Converting 2D drawings into 3D models will provide the foundational data needed to develop digital replicas of critical assets. These digital twins will allow the DoD to analyze performance, anticipate failures, and optimize readiness for legacy systems that remain integral to mission success.&nbsp;The expected outcomes include a transformative impact on the sustainment of aging platforms, the revitalization of the DIB through access to modern engineering data, and enhanced collaboration and innovation across the defense ecosystem. By reducing reliance on manual conversion processes, accelerating manufacturing timelines, and improving access to manufacturable models, this project will contribute to the DoD’s mission readiness while positioning the DIB to adapt to the evolving demands of national security.

## Objective

The end state of this project is to develop and deploy an AI-driven 2D-to-3D conversion system that enhances the sustainment and operational readiness of legacy systems within the Department of Defense (DoD). This system will address the critical challenges of maintaining aging equipment by leveraging AI technologies to convert outdated 2D engineering drawings into precise 3D models. These 3D models will form the foundation for advanced manufacturing, predictive maintenance, and digital twin creation, ultimately strengthening the Defense Industrial Base (DIB) and improving readiness across the DoD.&nbsp;In Phase I, the project will establish the feasibility of AI-based 2D-to-3D conversion through a detailed evaluation of the current state of the art, definition of a training corpus, and proof-of-concept testing on representative legacy drawings. The focus will be on determining the data requirements, technical gaps, and opportunities for leveraging secure data sharing through existing blockchain infrastructure.&nbsp;In Phase II, the project will develop a scalable prototype system integrating advanced AI algorithms tailored to the unique needs of engineering drawings. This system will be capable of converting complex and diverse 2D schematics into accurate, manufacturable 3D models. The system will also integrate into existing secure data sharing repositories to enable collaboration across the DIB, enhancing supply chain visibility and resilience.&nbsp;By Phase III, the fully operational system will be deployed at scale across the DoD and DIB. It will support the sustainment of legacy systems by enabling rapid part prototyping, reducing manufacturing lead times, and fostering innovation in maintenance and lifecycle management. Advanced features, such as digital twin creation, will further enhance readiness by enabling predictive analytics and efficient resource allocation. The system’s integration into existing systems will ensure secure collaboration and data integrity, aligning with DoD imperatives for operational resiliency and supply chain modernization.&nbsp;This project will result in a transformative capability for the DoD, addressing critical gaps in sustaining aging systems, bolstering the DIB’s ability to meet demand, and enabling a secure, collaborative approach to 21st-century defense logistics.

## Phase I

The primary focus of Phase I is to assess the current state of the art in AI tools for converting 2D engineering drawings into 3D CAD/CAM models and to determine the data requirements and feasibility of creating a robust AI-driven solution. This exploratory phase will evaluate existing technologies, define the corpus size and quality necessary for training AI models, and establish foundational requirements for subsequent development.&nbsp;Scope and Tasks:&nbsp;State-of-the-Art Analysis:<ul><li>Conduct a comprehensive review of existing AI tools, such as image-to-3D services (e.g., Luma AI, 3DFY AI), and evaluate their capabilities and limitations for precise engineering conversions.</li><li>Compare traditional CAD software with AI-assisted features to identify semi-automated approaches that could inform AI algorithm development.</li><li>Analyze industry-specific tools and custom development options for insights into engineering-specific requirements.</li></ul>&nbsp;Corpus Development Strategy:<ul><li>Define the type, size, and quality of 2D-to-3D data corpus required to train AI models effectively. This includes identifying DoD legacy systems with outdated 2D drawings as potential sources of data.</li><li>Partner with relevant stakeholders to gather representative samples of legacy 2D drawings and existing 3D CAD models for preliminary analysis.</li><li>Address challenges related to data security, access, and intellectual property, ensuring alignment with existing blockchain-secured data sharing framework.</li></ul>&nbsp;Proof-of-Concept Feasibility Study:<ul><li>Perform initial tests using existing AI tools on a small subset of legacy 2D drawings to assess conversion accuracy and detail retention.</li><li>Document findings and identify key gaps in current technologies that will guide algorithm development in Phase II.</li></ul>&nbsp;Deliverables:<ul><li>A detailed report on the state of the art, highlighting existing tools, their limitations, and potential enhancements.</li><li>A comprehensive plan for corpus development, including data requirements and security considerations.</li><li>A feasibility assessment of using AI for 2D-to-3D conversion, outlining technical challenges and opportunities.</li><li>SySML documentation of the complete effort.</li></ul>

## Phase II

Building on Phase I findings, Phase II will focus on the development and refinement of preliminary AI algorithms for 2D-to-3D conversion. This phase will also involve creating a scalable prototype system capable of handling diverse and complex engineering drawings, with integration into a blockchain-secured repository for testing and validation.&nbsp;Scope and Tasks:&nbsp;AI Algorithm Development:<ul><li>Design and implement preliminary AI algorithms to convert 2D engineering drawings into 3D models, focusing on accuracy and manufacturability.</li><li>Train the algorithms using the corpus defined in Phase I, ensuring the inclusion of complex geometries, multi-layered assemblies, and diverse drawing formats.</li></ul>&nbsp;Prototype System Development:<ul><li>Develop a scalable prototype system capable of processing a range of legacy 2D drawings efficiently.</li><li>Integrate the system with existing blockchain framework to enable secure storage and sharing of 3D models.</li><li>Incorporate user-friendly interfaces and tools to facilitate collaboration with Defense Industrial Base (DIB) partners.</li></ul>&nbsp;Operational Testing:<ul><li>Conduct rigorous testing of the prototype system using a diverse dataset of legacy 2D drawings from DoD systems.</li><li>Validate the accuracy, detail retention, and manufacturability of the generated 3D models through collaboration with DIB manufacturers.</li></ul>&nbsp;Feedback and Iteration:<ul><li>Gather feedback from end-users, including DoD engineers and DIB partners, to refine the algorithms and system capabilities.</li><li>Address identified gaps to improve system performance and usability.</li></ul>&nbsp;Deliverables:<ul><li>A functional prototype system capable of converting legacy 2D drawings into accurate 3D models.</li><li>Initial integration of the system into existing blockchain-secured repository for operational use.</li><li>Technical documentation and user guides for testing and feedback purposes.</li><li>A report on testing outcomes, system performance, and recommendations for Phase III deployment.</li><li>SySML documentation of the complete effort.</li></ul>

## Phase III: Dual Use Applications

The Phase III effort will transition the prototype system into a fully operational solution deployed across the DoD and DIB. This phase will focus on scalability, advanced features such as digital twin integration, and widespread adoption to enhance the resiliency of the Defense Industrial Base and sustainment of legacy systems.&nbsp;Scope and Tasks:&nbsp;Full-Scale System Deployment:<ul><li>Deploy the AI-driven 2D-to-3D conversion system across DoD platforms, integrating it within existing secure data-sharing infrastructure.</li><li>Scale the system to accommodate large volumes of legacy 2D drawings and expand its capabilities to include real-time processing.</li></ul>&nbsp;Advanced Features and Digital Twin Integration:<ul><li>Enhance the system to generate digital twins of legacy systems, enabling predictive maintenance, readiness assessments, and lifecycle management.</li><li>Develop analytics tools to support operational planning and decision-making based on digital twin insights.</li></ul>&nbsp;Operational Support and Training:<ul><li>Provide training programs for DoD personnel and DIB partners to ensure effective use of the system.</li><li>Establish long-term support mechanisms to address user needs and ensure system reliability.</li></ul>&nbsp;Impact Evaluation:<ul><li>Evaluate the system’s impact on sustainment efficiency, operational readiness, and cost savings.</li><li>Identify opportunities for further innovation and integration into other DoD programs.</li></ul>&nbsp;Deliverables:<ul><li>A fully operational system deployed across DoD and DIB, with secure access to a repository of 3D models and digital twins.</li><li>Comprehensive training and support materials to ensure widespread adoption and effective use.</li><li>A final report detailing system impact, scalability, and potential for further enhancements</li><li>SySML documentation of the complete effort.</li></ul>

PDF.html

Please provide the text you would like summarized.

Using AI to Create Multi-Dimensional digital Models from legacy drawings

STTR Opportunity Analysis

SBIR Documents

Related SBIR/STTR Opportunities

Opportunity Snapshot

Key Dates