Abstract
Decades after being replaced with digital platforms, analog computing has recently experienced a surge, capitalizing on developments in metamaterials and novel fabrication techniques. In the mechanical domain, much of the work has focused on quasi-static, bistable or origami-inspired materials which execute basic logic. In pursuit of more complex tasks, wave-based mechanical computers, which operate by imparting spatial transformations on impinging acoustic excitations, have gained traction owing to their ability to directly encode the input in its mechanical form, bypassing analog-to-digital conversion. Through guided wave scattering in tuned elastic media, these systems can undertake high-order computations such as differentiation and convolution or instigate neuromorphic intelligence by realizing physical neural architectures. While promising, the overarching constraints on elastic wave propagation including reciprocity, transmission symmetry, and preset dispersion patterns, have inherently limited these systems to single-task configurations. As such, their inability to re-adapt to new information or concurrently perform multiple tasks (i.e., compute in parallel) represents a major hindrance to advancing conceptual mechanical devices with broader computational capabilities. In this talk, I will provide an overview of our recent efforts to model and attain reconfigurable neuromorphic metasurfaces that perform multi-class classifications which can be altered on demand via different geometric knobs. Following which, I will discuss our attempts to simultaneously process independent computational tasks within the same architected structure via frequency multiplexing. By breaking time invariance in an array of tunable resonant cells, multiple frequency-shifted beams are self-generated which absorb notable energy amounts from the fundamental signal. The onset of these tunable harmonics enables distinct computational tasks to be assigned to different independent “channels,” effectively allowing a wave-based mechanical computer to multitask.
About the Speaker
Mostafa Nouh is an Associate Professor in the Department of Mechanical and Aerospace Engineering at the University at Buffalo (SUNY) and the director of the Sound and Vibrations Laboratory. He obtained his PhD degree (2013) in Mechanical Engineering from the University of Maryland, College Park. From 2013-2015, he served as a postdoctoral research associate at the Smart Materials and Structures Research Center at the University of Maryland. His research focuses on structural dynamics and acoustics with applications in phononic crystals and metamaterials, thermoacoustic energy generation, and non-reciprocal mechanics. He is a fellow of the American Society of Mechanical Engineers (ASME), a recipient of the NSF CAREER award, ASME’s Gary Anderson Early Achievement award, and the University at Buffalo’s Young Investigator and Meyerson awards. He has published over 60 journal papers and 4 book chapters, and serves as an Associate Editor of the ASME Journal of Vibration and Acoustics.
Host: Professor Sam Tawfick