Results 221 to 230 of about 1,658,125 (308)

Deep Learning Approaches for Classifying Crack States With Overload and Predicting Fatigue Parameters in a Titanium Alloy

open access: yesAdvanced Intelligent Systems, EarlyView.
This study proposes a deep learning approach to evaluate the fatigue crack behavior in metals under overload conditions. Using digital image correlation to capture the strain near crack tips, convolutional neural networks classify crack states as normal, overload, or recovery, and accurately predict fatigue parameters.
Seon Du Choi   +5 more
wiley   +1 more source

Deep Reinforcement Learning Approaches for Sensor Data Collection by a Swarm of UAVs

open access: yesAdvanced Intelligent Systems, EarlyView.
This article presents four decentralized reinforcement learning algorithms for autonomous data harvesting and investigates how collaboration improves collection efficiency. It also presents strategies to minimize training times by improving model flexibility, enabling algorithms to operate with varying number of agents and sensors.
Thiago de Souza Lamenza   +2 more
wiley   +1 more source

ResearchConnect: An AI‐Powered Platform for Interdisciplinary Research Team Formation and Ideation Development

open access: yesAdvanced Intelligent Systems, EarlyView.
ResearchConnect is an AI‐powered platform that automates researcher profiling, interdisciplinary team formation, and early‐stage research ideation. By extracting keywords from papers and web sources, it quickly clusters researchers into coherent teams and generates collaborative ideas using large language models. Validation on NSF‐funded projects shows
Akshay Vilas Jadhav   +2 more
wiley   +1 more source

Economy of Touch : Task‐Driven Information Selection in Electrical Impedance Tomography‐based Tactile Robotic Sensing

open access: yesAdvanced Intelligent Systems, EarlyView.
Electrical impedance tomography (EIT) tactile skins enable multiplexed measurements that trade sensing speed against information richness. This work introduces an economy‐of‐touch framework that treats tactile sensing as an information‐budgeting problem.
Xiaoxian Xu, David Hardman, Fumiya Iida
wiley   +1 more source

Disentangling Aleatoric and Epistemic Uncertainty in Physics‐Informed Neural Networks: Application to Insulation Material Degradation Prognostics

open access: yesAdvanced Intelligent Systems, EarlyView.
Physics‐Informed Neural Networks (PINNs) provide a framework for integrating physical laws with data. However, their application to Prognostics and Health Management (PHM) remains constrained by the limited uncertainty quantification (UQ) capabilities.
Ibai Ramirez   +4 more
wiley   +1 more source

Home - About - Disclaimer - Privacy