ABSTRACT

This Boosting Research Ideas for Transformative and Equitable Advances in Engineering (BRITE) Fellow grant will establish a transformative data-driven design framework enabled by artificial intelligence (AI) for the real-time digital design and fabrication of programmable material systems (PMS). PMS are emerging architectural structures made of smart materials that are responsive to external stimuli (e.g., stress, thermal inputs, chemical changes, light, and magnetic fields) and that can be programmed to transform between multiple functional states. PMS have far-reaching, societally impactful applications, including surgical robots, (bio)sensors, deployable satellites, mechanical computing, and water and energy harvesting. The design of PMS is still in its infancy, however, due to the complex underlying physics and high dimensionality associated with the design of spatially varying materials, architectures, and stimuli. To address these challenges, this project seeks to integrate disruptive technologies across the multidisciplinary domains of design, mechanics, manufacturing, materials, and data science to create a new AI-enabled PMS digital design platform. In collaboration with Minority Serving Institutions (MSIs), research results will be integrated into AI literacy programs and activities for K-12 and college students. A wide range of diversity, equity, and inclusion activities will also be accomplished, with emphasis on mentoring and collaboration with junior faculty from underrepresented groups and enhancing access to STEM pathways for underrepresented minority students.

The research objective of this project is to establish a novel data-driven design framework called ALGO (Acquire-Learn-Generate-Optimize) that will accelerate the co-design of materials (M), architectures (A), and stimuli (S) in programmable material systems (PMS). The specific goals are to: 1) Create a shared PMS data resource to bridge knowledge gaps across multiple disciplines and domains; 2) Develop novel statistical and AI-based learning techniques to understand complex M-A-S interactions and derive transferrable PMS design rules; and 3) Employ a ?building block? approach to create multiscale design strategies that combine machine learning with topology optimization to achieve superior computational efficiency and unprecedented performance for real-time PMS digital design. This research will provide a paradigm shift that transforms existing techniques limited to the design of single-material periodic structures into scalable data-driven design of programmable multi-material systems with heterogenous materials and topological architectures. While the PMS design testbeds used in this research will be focused on Shape Transformation, Wave Guiding, and Surface Engineering, the AI-enhanced learning and design automation techniques developed here will benefit a wide range of physics-driven science and engineering domains. Exploiting heterogeneity and programmability in material systems through intelligent design will have long-lasting impacts on US competitiveness in developing innovative, lightweight, portable, economic, and sustainable products.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Please report errors in award information by writing to: awardsearch@nsf.gov.