*This was a HYBRID Event with in-person attendance in Levine 307 and Virtual attendance via Zoom
Our research employs artificial intelligence techniques that seek to automate the main time/cost drivers of engineered-systems design. The features of a system inform the form, function and behavior of the resulting concept that can be subsequently created using traditional manufacturing/additive manufacturing methods. While there exists a wide range of computer aided design tools that seek to generate 3D design concepts, they are primarily parametric in nature and rely extensively on domain expertise, which may not always be readily available. Grants from the National Science Foundation (NSF) and the Defense Advanced Research Projects Agency (DARPA) have enabled our research team to explore the use of Deep Generative Design methods such as Generative Adversarial Networks (GANs) to generate 3D representations of design concepts. However, there is more to a design than simply its 3D form, as the design must perform a function and operate in an environment where its behavior may/may not perform as intended. Towards this end, our research group has proposed liking the AI-generation of a design, with the automatic evaluation of its function and behavior using physics-based simulation engines. The end result is a physics-informed design that has the potential to be realized through techniques such as additive manufacturing.