A Distributed Multi-Sensor Machine Learning Approach to Earthquake Early Warning

@inproceedings{Fauvel2020ADM,
  title={A Distributed Multi-Sensor Machine Learning Approach to Earthquake Early Warning},
  author={Kevin Fauvel and Daniel Balouek-Thomert and Diego Melgar and Pedro Silva and Anthony Simonet and Gabriel Antoniu and Alexandru Costan and V{\'e}ronique Masson and M. Parashar and Ivan Rodero and Alexandre Termier},
  booktitle={AAAI Conference on Artificial Intelligence},
  year={2020}
}
Our research aims to improve the accuracy of Earthquake Early Warning (EEW) systems by means of machine learning. EEW systems are designed to detect and characterize medium and large earthquakes before their damaging effects reach a certain location. Traditional EEW methods based on seismometers fail to accurately identify large earthquakes due to their sensitivity to the ground motion velocity. The recently introduced high-precision GPS stations, on the other hand, are ineffective to identify… 
View via Publisher
ojs.aaai.org
22 Citations
Highly Influential Citations
1
Background Citations
11
Methods Citations
5

Figures and Tables from this paper

22 Citations

Deep-learning seismology

Seismic waves from earthquakes and other sources are used to infer the structure and properties of Earth’s interior. The availability of large-scale seismic datasets and the suitability of…

EARTHQUAKE ALERT SYSTEM BASED ON IOT TECHNOLOGY

It is impossible to know where and when a seismic event can happen, thus to alleviate damages in infrastructure and human lives, is the early detection where a real-time architecture and an efficient communication becomes a requirement.

EARTHQUAKE ALERT SYSTEM BASED ON IOT TECHNOLOGY Under the Guidance of P.RAJA

It is impossible to know where and when a seismic event can happen, thus to alleviate damages in infrastructure and human lives, is the early detection where a real-time architecture and an efficient communication becomes a requirement.

Towards a semantic model for IoT-based seismic event detection and classification

An ontology model for describing the seismic domain is introduced, through the data collection from sensors, to gather seismic signals aimed at the seismic event recognition, and a machine learning-based classification has been accomplished to identify seismic events.

Recent Advances in Internet of Things Solutions for Early Warning Systems: A Review

The aim of the paper is to describe the adopted IoT architectures, define the constraints and the requirements of an Early Warning system, and systematically determine which are the most used solutions in the four use cases examined.

Harnessing the Computing Continuum for Urgent Science

An Early Earthquake Warning workflow, which combines data streams from geo-distributed seismometers and high-precision GPS stations to detect large ground motions, is proposed as a system stack that can enable the fluid integration of distributed analytics across a dynamic infrastructure spanning the computing continuum.

Approaches and Applications of Early Classification of Time Series: A Review

A systematic review of the current literature on early classification approaches for time series data, along with their potential applications, suggests some promising directions for further work in this area.

XCM: An Explainable Convolutional Neural Network for Multivariate Time Series Classification

XCM is a new compact convolutional neural network which extracts information relative to the observed variables and time directly from the input data, enabling a good generalization ability on both large and small datasets and allowing the full exploitation of a faithful post hoc model-specific explainability method.

Explainable AI for Car Crash Detection using Multivariate Time Series

  • L. TronchinR. Sicilia P. Soda
  • Computer Science
    2021 IEEE 20th International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC)
  • 2021
This paper presents the application and evaluation of three eXplainable Artificial Intelligence methods in a real-world multimodal task of anomaly detection on telematics data and deals with the challenge of explaining Multivariate Time Series and of translating methods designed for images to this domain.

Facilitating Data Discovery for Large-scale Science Facilities using Knowledge Networks

This paper develops the collaborative knowledge-aware graph attention network (CKAT) recommendation model, which leverages graph neural networks to explicitly encode the collaborative signals through propagation and combine them with knowledge associations, and integrates aknowledge-aware neural attention mechanism to enable the CKAT to pay more attention to key information while reducing irrelevant noise.

References

SHOWING 1-10 OF 28 REFERENCES

Machine Learning Seismic Wave Discrimination: Application to Earthquake Early Warning

This study trains a generative adversarial network to learn the characteristics of first‐arrival earthquake P waves, using 300,000 waveforms recorded in southern California and Japan, and demonstrates that GANs can discover a compact and effective representation of seismic waves, which has the potential for wide applications in seismology.

Convolutional neural network for earthquake detection and location

This work leverages the recent advances in artificial intelligence and presents ConvNetQuake, a highly scalable convolutional neural network for earthquake detection and location from a single waveform, and applies it to study the induced seismicity in Oklahoma, USA.

A wireless mesh sensing network for early warning

On robust and reliable automated baseline corrections for strong motion seismology

A proposed automated baseline correction algorithm using only accelerometer data is analyzed and it is found that the error incurred from automated baseline corrections that rely on seismic data alone is complex and can be large in both the time and frequency domains of interest in seismological and engineering applications.

Earthquake detection through computationally efficient similarity search

FAST detected most (21 of 24) cataloged earthquakes and 68 uncataloged earthquakes in 1 week of continuous data from a station located near the Calaveras Fault in central California, achieving detection performance comparable to that of autocorrelation, with some additional false detections.

Earthquake Early Warning and Tsunami Warning of JMA for the 2011 off the Pacific Coast of Tohoku Earthquake

The 2011 off the Pacific coast of Tohoku Earthquake (Mw9.0) occurred on March 11, 2011, caused strong ground motion around northeastern Japan, and generated devastating tsunami, which killed more…

Quantifying the Value of Real‐Time Geodetic Constraints for Earthquake Early Warning Using a Global Seismic and Geodetic Data Set

Geodetic earthquake early warning (EEW) algorithms complement point‐source seismic systems by estimating fault‐finiteness and unsaturated moment magnitude for the largest, most damaging earthquakes.…

Earthquake Early Warning: Advances, Scientific Challenges, and Societal Needs

The wealth of information about EEW uses and user needs must be employed to focus future developments and improvements in EEW systems, and the development of new, potentially transformative ideas and methodologies that could change how the authors provide alerts in the future are discussed.

Earthquake magnitude calculation without saturation from the scaling of peak ground displacement

GPS instruments are noninertial and directly measure displacements with respect to a global reference frame, while inertial sensors are affected by systematic offsets—primarily tilting—that adversely…

Geospatial Cyberinfrastructure: Past, present and future