The MIT researchers have developed a new approach using artificial intelligence (AI) to determine the internal properties of materials by observing their surface. By training an AI model with a large amount of data on surface measurements and their corresponding interior properties, the researchers were able to generate a system that can make reliable predictions about the material’s interior. This method is noninvasive and eliminates the need for expensive and bulky equipment like X-rays.
The AI model was trained using different types of materials, including composites, which are commonly used in new airplanes. It can also be applied to determine various properties, such as stress, strain, fluid fields, and magnetic fields. The technique even works for complex biological tissues where the behavior is not fully understood. By collecting enough data, the model can make accurate predictions even without a complete understanding of the material’s characteristics.
The researchers believe that this method has broad applicability across different engineering disciplines, including solid mechanics and fluid dynamics. It can be used to improve the design of materials for soft robotics applications, where the internal properties of the materials are often unknown. The method allows researchers to make predictions about what’s happening inside the material and design better materials accordingly.
In terms of practical applications, this new approach can revolutionize inspections in industries like aerospace. Rather than testing the entire airplane using expensive methods like X-rays, engineers can use the AI model to guide them on where to focus their inspections. This reduces costs and improves efficiency.
The research was supported by various organizations, including the U.S. Army Research Office, the Air Force Office of Scientific Research, Google Cloud platform, and the MIT Quest for Intelligence. The method is freely available to anyone through the website GitHub.
Overall, this new AI-based approach allows engineers to gain insight into the internal properties of materials by simply observing their surfaces. It has the potential to revolutionize various industries and improve the design and manufacturing processes for materials.