Séminaire LTDS (en ligne) : Reconfigurable Metastructures – From Wave Control to Mechano-Intelligence

Présenté par : Kon-Well Wang
Stephen P. Timoshenko Professor of Mechanical Engineering, University of Michigan, Ann Arbor, MI, USA

Résumé + biographie de l'orateur au format pdf

In recent years, the concept of mechanical metastructures developed based on nature-inspired modular architectures has been explored to create advanced engineering systems. For example, inspired by the observation that some of skeletal muscle's intriguing macroscale functionalities result from the assembly of nanoscale cross-bridge constituents with metastability, the idea of synthesizing structures from the integration of mechanical metastable modules has been pursued. In another example, inspired by the physics behind the plant nastic movements and the rich designs of origami folding, a class of metastructures is created building on the innovation of fluidic-origami modular elements. Overall, the metastructure modules are designed to be reconfigurable in their shape, mechanical properties, multi-stability features, and dynamic characteristics, so to produce synergistic and intriguing functionalities at the system level, such as programmable phononic bandgap control and nontraditional wave steering. More recently, with the rapid advances in high-performance intelligent systems, we are witnessing a prominent demand for the next generation of metastructures to have much more built-in intelligence and autonomy. An emerging direction is therefore to pioneer and harness the metastructures’ high dimensionality, multi-stability, and nonlinearity for mechano-intelligence via physical computing. That is, we aim to concurrently embed computing power and functional intelligence, such as observation, learning, memorizing, decision-making and execution, directly in the mechanical domain, advancing from conventional systems that solely rely on add-on electronics to achieve intelligence. This presentation will highlight some of these recent advancements in reconfigurable metastructures, from adaptive wave control to self-learning-self-tuning structural intelligence.