Event-triggered Natural Hazard Monitoring with Convolutional Neural Networks on the Edge

TL;DR

This paper presents a low-power, event-triggered wireless sensor network utilizing convolutional neural networks for real-time natural hazard monitoring, demonstrating effective hazard detection and classification on resource-constrained edge devices.

Contribution

It introduces an energy-efficient, event-triggered sensor system with CNN-based classification optimized for low-power microcontrollers, enabling real-time hazard detection at the edge.

Findings

System has been operational since 2018 at the Matterhorn for hazard monitoring.

CNN inference is optimized for low-power devices through quantization and pipelining.

The approach reduces false positives and improves response time in hazard detection.

Abstract

In natural hazard warning systems fast decision making is vital to avoid catastrophes. Decision making at the edge of a wireless sensor network promises fast response times but is limited by the availability of energy, data transfer speed, processing and memory constraints. In this work we present a realization of a wireless sensor network for hazard monitoring based on an array of event-triggered single-channel micro-seismic sensors with advanced signal processing and characterization capabilities based on a novel co-detection technique. On the one hand we leverage an ultra-low power, threshold-triggering circuit paired with on-demand digital signal acquisition capable of extracting relevant information exactly and efficiently at times when it matters most and consequentially not wasting precious resources when nothing can be observed. On the other hand we utilize machine-learning-based classification implemented on low-power, off-the-shelf microcontrollers to avoid false positive warnings and to actively identify humans in hazard zones. The sensors' response time and memory requirement is substantially improved by quantizing and pipelining the inference of a convolutional neural network. In this way, convolutional neural networks that would not run unmodified on a memory constrained device can be executed in real-time and at scale on low-power embedded devices. A field study with our system is running on the rockfall scarp of the Matterhorn H\"ornligrat at 3500 m a.s.l. since 08/2018.