New AI “copilot” helps engineers design better structures

MIT researchers have developed an AI "copilot" system that helps engineers design better structures. This system utilizes topology optimization, an advanced algorithmic design technique commonly used in the automotive and aerospace industries. However, the process often takes a significant amount of computation and time, resulting in impractical designs or other issues that designers may not have considered.

The new AI copilot system acts as a collaborative partner during the design process. It presents regions to the human designer for feedback, enabling them to make changes and influence the evolution of the design. This approach is compared to having a design copilot, as it brings together the strengths of algorithmic and human design methods.

According to the researchers, the human-in-the-loop approach improved manufacturability and linear buckling load by 39 percent while only slightly increasing the total design time. This collaborative system aims to make topology optimization more accessible and manufacturable for a broader range of engineering disciplines, including buildings and bridges. By integrating human input, the researchers hope to increase trust in the final designs and facilitate the acceptance of high-performing structures in the real world.