Pingchuan Yang

ORCID: 0009-0005-2347-6903
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Indoor and Outdoor Localization Technologies
  • GaN-based semiconductor devices and materials
  • Robotics and Sensor-Based Localization
  • Metal and Thin Film Mechanics
  • Welding Techniques and Residual Stresses
  • Advanced Surface Polishing Techniques
  • Metallurgy and Material Forming
  • Target Tracking and Data Fusion in Sensor Networks
  • Non-Destructive Testing Techniques
  • Laser and Thermal Forming Techniques
  • ZnO doping and properties
  • Ga2O3 and related materials
  • Machine Learning and Data Classification
  • Fatigue and fracture mechanics
  • Neural Networks and Applications
  • Online and Blended Learning
  • Mechanical Behavior of Composites
  • Copper-based nanomaterials and applications
  • Distributed Control Multi-Agent Systems
  • Aluminum Alloy Microstructure Properties
  • Electronic and Structural Properties of Oxides
  • Advanced Welding Techniques Analysis
  • Gas Dynamics and Kinetic Theory
  • Machine Learning in Healthcare
  • Additive Manufacturing Materials and Processes

Anhui University
2025

National University of Singapore
2013

Advanced Semiconductor Engineering (Taiwan)
2010-2011

Modern foundation models often undergo iterative ``bootstrapping'' in their post-training phase: a model generates synthetic data, an external verifier filters out low-quality samples, and the high-quality subset is used for further fine-tuning. Over multiple iterations, model's performance improves--raising crucial question: how should total budget on generation training be allocated across iterations to maximize final performance? In this work, we develop theoretical framework analyze...

10.48550/arxiv.2501.18962 preprint EN arXiv (Cornell University) 2025-01-31

With the rapid advancements in artificial intelligence (AI), 5G technology, and robotics, multi-sensor fusion technologies have emerged as a critical solution for achieving high-precision localization mobile robots operating within dynamic unstructured environments. This study proposes novel hybrid framework that combines Extended Kalman Filter (EKF) Recurrent Neural Network (RNN) to address challenges such sensor frequency asynchrony, drift accumulation, measurement noise. The EKF provides...

10.1038/s41598-025-90492-4 article EN cc-by-nc-nd Scientific Reports 2025-03-10

Abstract Accurate position information is crucial for unmanned aerial vehicles (UAV) to execute tasks. To balance the contradiction between payload and localization accuracy of rotary UAVs, a cooperative method UAV swarms based on factor graph model studied in this paper. Each equipped with local model. A distributed model-solving method, AGA-Gauss–Newton conjugate gradient (GNCG), which combines an adaptive genetic algorithm improved GNCG algorithm, proposed. The issue falling into optimal...

10.1088/1361-6501/ad91d6 article EN Measurement Science and Technology 2024-11-22

High resolution X-ray diffractometry (HR-XRD), Photoluminescence, Raman spectroscopy, and Transmission electron microscope measurements are reported for GaN deposited on a conventional Si(111) substrate the {111} facets etched Si(100) substrate. HR-XRD reciprocal space mappings showed that GaN(0002) plane is tilted by about 0.63° ± 0.02° away from exposed Si{111} growth surface patterned substrate, while no observable tilt existed between planes The ratio of integrated intensities yellow to...

10.1063/1.4856275 article EN Journal of Applied Physics 2013-12-27

An investigation was conducted on electron beam-welded and additively manufactured joints a thick-walled titanium alloy utilizing in situ laser beam deposition welding techniques. The surface morphology, microstructural characteristics, mechanical properties of both joint types were comprehensively analyzed using stereomicroscopy, scanning microscopy (SEM), microhardness tensile strength testing, backscatter diffraction (EBSD) electron-beam-welded exhibited distinct fusion heat-affected...

10.3390/mi15111383 article EN cc-by Micromachines 2024-11-15
Coming Soon ...