Amir Behjat

Amir Behjat's research combines machine learning with physics-based modeling to solve engineering design and robotics problems, including predicting acoustic fields from drones, designing materials that suppress vibration, and coordinating teams of robots (swarms) in complex environments. Much of the work involves developing neural network methods, such as invertible neural networks and neuroevolution (using evolutionary algorithms to design neural networks), that incorporate physical principles rather than relying on data alone. This work spans applications from aerospace structures to lunar habitat safety and robot swarm tactics.

Publication output rose from 2017 through a peak around 2019, then has fluctuated at a lower and less consistent pace in recent years, with only a few outputs per year since 2022.

Generated by claude-sonnet-5 from public bibliographic data · Jul 20, 2026