Search results

The pedestrian dead reckoning (PDR) based on smartphones has been widely applied in continuous indoor positioning. However, when the position of the mobile phone and the walking patterns of the pedestrian are mixed, traditional PDR tends to become confused and thus degrade performance. To address this issue, this paper aims to propose an improved PDR scheme by focusing on gait pattern recognition and the impact of short-period but negative transitions on tracking.

The overall solution uses the inertial sensor integrated within the phone for positioning. A binary classifier-based change point detection algorithm is used to identify the transition points in pedestrian gait. Additionally, to enhance the accuracy of gait recognition, this paper presents a combined CNN-attention-based bi-directional long short-term memory(ABiLSTM) model, integrating convolutional neural networks (CNN), bi-directional long short-term memory (Bi-LSTM) and an attention mechanism, to recognize the current gait pattern. The outcomes of this gait pattern recognition are then applied to PDR. Based on distinct gait patterns, corresponding PDR strategies are devised to enable continuous tracking and positioning of pedestrians.

Through experimental verification, the CNN-ABiLSTM model achieves a gait recognition accuracy of 99.52% on the self-constructed data set. The pedestrian navigation estimation method proposed in this paper, which is based on gait recognition assistance, demonstrates a 32.56% improvement in accuracy over traditional positioning algorithms in multi-gait scenarios.

The improved PDR scheme algorithm significantly enhances the robustness and smoothness of pedestrian tracking, particularly during multiple gait transitions. This, in turn, provides strong support for the utilization of low-cost inertial sensors integrated within mobile phones for indoor positioning applications.

The purpose of this study was to validate the feasibility of the proposed microstructure-based model by comparing the simulation results with experimental data. The study also aimed to investigate the relationship between the orientation of graphite flakes and the failure behavior of the material under compressive loads as well as the effect of image size on the accuracy of stress–strain behavior predictions.

This paper presents a microstructure-based model that utilizes the finite element method (FEM) combined with representative volume elements (RVE) to simulate the hardening and failure behavior of ferrite-pearlite matrix gray cast iron under uniaxial loading conditions. The material was first analyzed using optical microscopy, scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDS) and X-ray diffraction (XRD) to identify the different phases and their characteristics. High-resolution SEM images of the undeformed material microstructure were then converted into finite element meshes using OOF2 software. The Johnson–Cook (J–C) model, along with a damage model, was employed in Abaqus FEA software to estimate the elastic and elastoplastic behavior under assumed plane stress conditions.

The findings indicate that crack initiation and propagation in gray cast iron begin at the interface between graphite particles and the pearlitic matrix, with microcrack networks extending into the metal matrix, eventually coalescing to cause material failure. The ferritic phase within the material contributes some ductility, thereby delaying crack initiation.

This study introduces a novel approach by integrating microstructural analysis with FEM and RVE techniques to accurately model the hardening and failure behavior of gray cast iron under uniaxial loading. The incorporation of high-resolution SEM images into finite element meshes, combined with the J–C model and damage assessment in Abaqus, provides a comprehensive method for predicting material performance. This approach enhances the understanding of the microstructural influences on crack initiation and propagation, offering valuable insights for improving the design and durability of gray cast iron components.