Engineers at the University of Maryland (UMD) have created a model that combines machine learning and collaborative robotics to address challenges in designing materials for wearable technology. This accelerated method of creating aerogel materials could automate the design process for new materials, making the process more efficient.
Despite the simple nature of aerogels, the assembly line is complex and researchers rely on time-consuming experiments and experience-based approaches to explore different design options. To overcome these challenges, the research team integrated robotics, machine learning algorithms, and materials science expertise to accelerate the design of aerogels with programmable mechanical and electrical properties.
The prediction model developed by the team is capable of generating sustainable products with an impressive 95% accuracy rate. Po-Yen Chen, the lead researcher, emphasized that their workflow combining robotics and machine learning not only improves data quality but also assists researchers in navigating the intricate design space of wearable technology.
The team’s aerogels were created using conductive titanium nanosheets, cellulose, and gelatine, resulting in strong and flexible materials. They believe their approach can be expanded to other applications in aerogel design, such as green technologies for oil spill cleanup, sustainable energy storage, and thermal energy products like insulating windows.
Collaborator Eleonora Tubaldi expressed excitement about the combination of approaches putting them at the forefront of materials design with customizable properties. The team envisions utilizing this scale-up production platform to design aerogels with unique mechanical, thermal, and electrical properties for challenging working environments.