Geomagnetic-based indoor positioning has drawn a great attention from academia and industry due to its advantage of being operable without infrastructure support and its reliable signal characteristics. However, it must overcome the problems of ambiguity that originate with the nature of geomagnetic data. Most studies manage this problem by incorporating particle filters along with inertial sensors. However, they cannot yield reliable positioning results because the inertial sensors in smartphones cannot precisely predict the movement of users. There have been attempts to recognize the magnetic sequence pattern, but these attempts are proven only in a one-dimensional space, because magnetic intensity fluctuates severely with even a slight change of locations. This paper proposes accurate magnetic indoor localization using deep learning (AMID), an indoor positioning system that recognizes magnetic sequence patterns using a deep neural network. Features are extracted from magnetic sequences, and then the deep neural network is used for classifying the sequences by patterns that are generated by nearby magnetic landmarks. Locations are estimated by detecting the landmarks. AMID manifested the proposed features and deep learning as an outstanding classifier, revealing the potential of accurate magnetic positioning with smartphone sensors alone. The landmark detection accuracy was over 80% in a two-dimensional environment.

A magnetic indoor positioning system utilizes distorted geomagnetic fields indoors to estimate locations. However, existing magnetic positioning methods have difficulties in positioning, especially in a wide space because of the ambiguity of magnetic data. Multiple magnetic sensors should work in collaboration to yield a high accuracy, hindering the techniques from being used by mobile devices such as smartphones. To address this problem, we propose a new magnetic indoor positioning method using deep neural network. Features are extracted from magnetic sequences, and then the deep neural network is used for classifying features of magnetic landmarks detected from a dense magnetic map. The magnetic map is constructed by a robot. The proposed method achieved over 80% accuracy in a two-dimensional environment.

We study the configurational and dynamic properties of a long polymer chain during the process of penetration of the chain through the thin and porous membrane. The mobility of a polymer in this state depends strongly on the topological restrictions imposed by the membrane. The largest mobility have configurations in which the polymer chain crosses the membrane only once and diffuses through a single pore from one side of the membrane to the other. This configuration may have small statistical weight compared to other configurations with low mobility in which the polymer chain crosses the membrane many times. The relative weight of these configurations depends also on the chemical potential of the membrane. We find the configurations which give the main contribution to the transport and calculate the overall permeability of the membrane for Zimm and Rouse mechanisms of polymer dynamics

Aeroengine blisks operate in a harsh working environment and are prone to low cycle fatigue (LCF) failure. The probabilistic LCF life prediction considering multiple uncertainties needs to be performed for reliability assessment. To consider the combined effects of heterogeneous uncertainties, this article employs a unified reliability assessment method by processing the uncertainties simultaneously. To overcome the extremely time-consuming limitation of probabilistic finite-element model simulation, this article develops an ensemble generalized constraint neural network (EGCNN)-based unified reliability assessment method. The developed EGCNN surrogate model can conduct efficient, accurate, interpretable, and robust reliability assessments with nonlinear fitting capability, knowledge interpretability, and premature avoidance ability. The developed EGCNN-based unified reliability assessment method can also be applied to other assets and failure mechanisms, providing a new reliability-based design optimization tool.

A home indoor positioning system (HIPS) is a positioning system that provides location information of mobile devices, such as smartphones, for location-based IoT applications in a home environment. This paper introduces methods and algorithms to construct an HIPS using Wi-Fi signals. In the proposed system, an intelligent mobile robot, such as a home vacuum robot, is utilized to automatically construct radio maps for the system. Along with radio maps, the system adopts new algorithms to provide highly accurate positioning services and provides interfaces to support variable location-based IoT applications.

This paper presents a feasibility study to diagnose faults in automotive safety components that are subjected to abnormal vibrations. Diagnosis targets six faults from different components that generate abnormal vibrations and faults during operation. Four deep learning approaches were developed and evaluated in terms of their suitability for embedding inside a vehicle. As a result, all four architectures were trained and executed on a Raspberry Pi to replicate the expected computational power of the embedded system.