My grandfather possessed an exceptional talent for understanding and repairing mechanical devices, which he used to support his modest shop on the outskirts of El Paso, Texas, a border town with a unique character.
A two-step approach is proposed to learn nonlinear Lagrangian reduced-order models (ROMs) of nonlinear mechanical systems directly from data, without requiring access to the full-order model operators. The method first learns a linear Lagrangian ROM via Lagrangian operator inference and then augments it with nonlinear terms learned using structure-preserving machine learning.
Understanding the interplay between symmetries, mechanical connections, and impacts in hybrid mechanical systems.
Gearbox fault diagnosis datasets facilitate testing new methods effectively.
提案された機構戦略に基づく一自由度同期放射運動を持つキリガミアルキメデス多面体の設計と変換可能性に関する革新的な解決策。
The author proposes the Prototype Matching Network (PMN) to enhance interpretability in mechanical fault diagnosis by combining prototype-matching with autoencoder. The PMN offers insights into classification logic, typical fault signals, and frequency contributions for better representation learning.
The author proposes a family of kirigami Archimedean polyhedrons based on spatial 7R linkages to achieve one-DOF radial transformations, enabling rich configurational changes. This innovative approach facilitates applications in aerospace exploration, architecture, and metamaterials.
The author proposes a two-stage approach to time-optimal point-to-point motion planning, combining fixed and variable time grids for computational manageability and avoidance of interpolation errors. The integration with an asynchronous NMPC update scheme facilitates online replanning.