1710 papers
This paper presents a large-scale weakly supervised machine learning framework that analyzes global telemetry from 500 vaccine refrigerators to jointly quantify human operational behaviors and thermal responses, enabling the accurate detection of six interpretable anomaly categories that compromise cold-chain integrity.
This paper presents and validates a unified dynamic model for the MD-LUFFY hyper-extension pneumatic actuator that accurately predicts both static deformation and dynamic response under varying material and pneumatic conditions by integrating nonlinear rubber properties, fiber constraints, and supply system dynamics.
This study develops and validates a trajectory-based digital twin simulation of fused deposition modeling (FDM) that accurately predicts residual stress and shape distortion in thin-walled complex PLA geometries, offering a predictive tool for optimizing printing parameters and improving manufacturability.
This study demonstrates that optimizing the enclosure geometry of a passive phase change material-based battery thermal management system significantly enhances thermal performance, reducing peak cell temperature and improving temperature uniformity for high-rate 18650 LiFePO4/Graphite battery packs.
This paper systematically analyzes quorum-sensing protocols for minimalist swarm robots, revealing that while anonymous protocols offer memory efficiency, identity-aware strategies with re-broadcasting provide superior accuracy and coordination, particularly in non-well-mixed environments.
This study introduces ASND-DEM, an adaptive framework that optimizes DEM resolution for photogrammetric point clouds in open-pit mining by dynamically adjusting grid density based on local point spacing and terrain roughness, achieving significant data compression while preserving critical geological features.
This paper presents and experimentally validates a decentralized, edge-enabled smart compaction module for paper waste that utilizes local sensors and automated volume reduction to optimize collection routes, increase bin capacity, and lower operational costs and carbon footprints.
This paper proposes a spiking-neural-network-based deep reinforcement learning method, featuring a Spiking Actor Network and an extended replay buffer, to achieve energy-efficient, autonomous, and coordinated locomotion control for a shape-memory-alloy-driven multi-legged soft robot.
This paper presents the design, development, and real-world validation of a low-cost, autonomous feed pusher robot for dairy barns that utilizes an Agilex tracked chassis, LiDAR and IMU sensors, and ROS 2 Navigation 2 to navigate uneven surfaces and avoid obstacles without requiring structural modifications to the facility.
This study utilizes ABAQUS-based numerical modeling and orthogonal tests to systematically analyze the effects of finite soil conditions and various parameters on earth pressure and excavation deformation, ultimately deriving predictive formulas and an optimization table for retaining structures in foundation pits adjacent to existing buildings.