Combined CNN and RNN Neural Networks for GPR Detection of Railway Subgrade Diseases

Vehicle-mounted ground-penetrating radar (GPR) has been used to non-destructively inspect and evaluate railway subgrade conditions. However, existing GPR data processing and interpretation methods mostly rely on time-consuming manual interpretation, and limited studies have applied machine learning methods. GPR data are complex, high...

Comparing multichannel mixed CNN-RNN to individual models for earthquake prediction

Earthquakes can do great harm to the environment and people's daily lives. Being able to predict an earthquake moments before it happens could therefore reduce harm and save human lives. Traditional methods have not been successful yet, but with the rise of techniques focused on deep learning, there is a growing interest to apply them to the...

Probabilistic Estimation of InSAR Displacement Phase Guided by Contextual Information and Artificial Intelligence

Phase unwrapping, also known as ambiguity resolution, is an underdetermined problem in which assumptions must be made to obtain a result in SAR interferometry (InSAR) time series analysis. This problem is particularly acute for distributed scatterer InSAR, in which noise levels can be so large that they are comparable in magnitude to the...

A 23-μW Keyword Spotting IC With Ring-Oscillator-Based Time-Domain Feature Extraction

This article presents the first keyword spotting (KWS) IC that uses a ring-oscillator-based time-domain processing technique for its analog feature extractor (FEx). Its extensive usage of time-encoding schemes allows the analog audio signal to be processed in a fully time-domain manner except for the voltage-to-time conversion stage of the...

A hybrid approach to implement the Digital Twin concept into a damage evolution prediction for composite structures

Improvement in the SHM of composite materials requires an enhanced understanding of the damage accumulation processes and helps in the way towards lighter, more optimized, and more sustainable aerospace structures. The Digital Twin concept has the potential to address this problem and may revolutionize the designing, certifying, maintaining, and...

Improving the performance of Recurrent Neural Networks for time series prediction by combining Long Short-Term Memory and Attention Long Short-Term Memory

Recurrent neural networks (RNNs) used in time series prediction are still not perfect in their predictions and improvements can still be made in the area. Most recently transformers have led to great improvements in the field of RNNs, however transformers can not be used on time series data, because the architecture of transformers does not...

Generalized Variant Support Vector Machine

With the advancement in information technology, datasets with an enormous amount of data are available. The classification task on these datasets is more time- and memory-consuming as the number of data increases. The support vector machine (SVM), which is arguably the most popular classification technique, has disappointing performance in...

Degradation Estimation and Prediction of Electronic Packages using Data Driven Approach

Recent trends in automotive electronics such as automated driving will increase the number and complexity of electronics used in safety-relevant applications. Applications in logistics or ridesharing will require a specific year of service rather than the conventional mileage usage. Reliable operations of the electronic systems must be...

Distributed FDI using Recurrent Neural Networks for Thruster Faults in Formation Flying Satellites

Certain Formation Flying missions rely on their orbital control thrusters to maintain the formation, making a Fault Detection and Isolation (FDI) system that is capable of detecting thruster faults very valuable. In addition, communication between satellites in a formation can be expensive. In this thesis, a distributed FDI approach was...

Enhancing crowd monitoring system functionality through data fusion: Estimating flow rate from wi-fi traces and automated counting system data

Crowd monitoring systems (CMSs) provide a state-of-the-art solution to manage crowds objectively. Most crowd monitoring systems feature one type of sensor, which severely limits the insights one can simultaneously gather regarding the crowd’s traffic state. Incorporating multiple functionally complementary sensor types is expensive. CMSs are...

Forecasting pedestrian movements using recurrent neural networks: An application of crowd monitoring data

Currently, effective crowd management based on the information provided by crowd monitoring systems is difficult as this information comes in at the moment adverse crowd movements are already occurring. Up to this moment, very little forecasting techniques have been developed that predict crowd flows a longer time period ahead. Moreover, most...

From Deterministic to Generative: Multimodal Stochastic RNNs for Video Captioning

Video captioning, in essential, is a complex natural process, which is affected by various uncertainties stemming from video content, subjective judgment, and so on. In this paper, we build on the recent progress in using encoder-decoder framework for video captioning and address what we find to be a critical deficiency of the existing...