Lightweight Edge Intelligence for Early Detection of Abnormal Machine Behaviour Using ESP32

Machines often show early signs of mechanical problems through unusual vibrations or changes in their normal operating behaviour. If these signs are not detected in time, they can lead to equipment failure, unexpected downtime, and increased maintenance costs. Traditional monitoring systems often rely on cloud-based processing or fixed threshold values, which may not always be suitable for affordable and resource-limited applications. This paper introduces TinyEdge-IoT, a proposed lightweight edge-intelligence framework for identifying abnormal machine behaviour using an ESP32 microcontroller. The system is designed around low-cost sensors, including the MPU6050, which can collect vibration and motion data from operating machinery. The collected data is processed directly on the ESP32 through a sequence of steps involving data collection, preprocessing, feature extraction, and lightweight anomaly detection. Instead of sending all sensor readings to a remote server, the proposed approach performs the main analysis locally. This can help reduce dependence on continuous internet connectivity and support faster responses to unusual machine behaviour. Wi-Fi and MQTT can be used to send alerts, record important events, and display monitoring information remotely. The paper also presents an evaluation plan to study the framework under normal and controlled abnormal operating conditions. Performance can be assessed using precision, recall, F1-score, false-alarm rate, processing time, memory usage, and energy consumption. Since the proposed system has not yet been implemented and experimentally tested, no measured performance results are claimed. The work focuses on the system architecture, its design considerations, and a methodology for future validation. TinyEdge-IoT aims to provide a practical starting point for affordable machine-condition monitoring in small motors, laboratories, and workshop environments.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-14
DOI
https://doi.org/10.5281/zenodo.24212155
Primary Topic
Machine Fault Diagnosis Techniques
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Lightweight Edge Intelligence for Early Detection of Abnormal Machine Behaviour Using ESP32

Rishab Raj, Ankit Kumar Sah, Bindu Thakral
Zenodo (CERN European Organization for Nuclear Research)
Machine Fault Diagnosis Techniques
article

Lightweight Edge Intelligence for Early Detection of Abnormal Machine Behaviour Using ESP32

Rishab Raj, Ankit Kumar Sah, Bindu Thakral
article en

Abstract

Machines often show early signs of mechanical problems through unusual vibrations or changes in their normal operating behaviour. If these signs are not detected in time, they can lead to equipment failure, unexpected downtime, and increased maintenance costs. Traditional monitoring systems often rely on cloud-based processing or fixed threshold values, which may not always be suitable for affordable and resource-limited applications. This paper introduces TinyEdge-IoT, a proposed lightweight edge-intelligence framework for identifying abnormal machine behaviour using an ESP32 microcontroller. The system is designed around low-cost sensors, including the MPU6050, which can collect vibration and motion data from operating machinery. The collected data is processed directly on the ESP32 through a sequence of steps involving data collection, preprocessing, feature extraction, and lightweight anomaly detection. Instead of sending all sensor readings to a remote server, the proposed approach performs the main analysis locally. This can help reduce dependence on continuous internet connectivity and support faster responses to unusual machine behaviour. Wi-Fi and MQTT can be used to send alerts, record important events, and display monitoring information remotely. The paper also presents an evaluation plan to study the framework under normal and controlled abnormal operating conditions. Performance can be assessed using precision, recall, F1-score, false-alarm rate, processing time, memory usage, and energy consumption. Since the proposed system has not yet been implemented and experimentally tested, no measured performance results are claimed. The work focuses on the system architecture, its design considerations, and a methodology for future validation. TinyEdge-IoT aims to provide a practical starting point for affordable machine-condition monitoring in small motors, laboratories, and workshop environments.

Zenodo (CERN European Organization for Nuclear Research)
Machine Fault Diagnosis Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon