Using sensor fusion and tinyML to detect fires
Using sensor fusion and tinyML to detect fires
Arduino Team — February 13, 2023
The damage and destruction caused by structural fires to people and the property itself is immense, which is why accurate and reliable fire detection systems are essential. As Nekhil R. notes in his article, current rule-based algorithms and simple sensor configurations can lead to reduced accuracy, showing the need for more robust systems.
This led Nekhil to design a solution that leverages sensor fusion and machine learning to make better predictions about the presence of flames. His project began by collecting environmental data including temperature, humidity and pressure from his Arduino Nano 33 BLE Sense's onboard sensor suite. He also labeled each Fire or No Fire sample using Edge Impulse Studio, which was used to generate spectral features from the three time-series sensor values. This information was then fed to a Keras neural network that had been configured to perform the classification, resulting in an overall accuracy of 92.86% when run on real-world test samples.
>
Confident in his now trained model, Nekhil deployed his model as an Arduino library on the Nano 33 BLE Sense. The Nano sends a message through its UART pins to a standby ESP8266-01 board when a fire has been detected. And in turn, the ESP8266 triggers an IFTTT webhook to alert the user via email.
If you want to know more about the construction of this fire recognition system, you will find many details on the project page.
Arduino Team — February 13, 2023
The damage and destruction caused by structural fires to people and the property itself is immense, which is why accurate and reliable fire detection systems are essential. As Nekhil R. notes in his article, current rule-based algorithms and simple sensor configurations can lead to reduced accuracy, showing the need for more robust systems.
This led Nekhil to design a solution that leverages sensor fusion and machine learning to make better predictions about the presence of flames. His project began by collecting environmental data including temperature, humidity and pressure from his Arduino Nano 33 BLE Sense's onboard sensor suite. He also labeled each Fire or No Fire sample using Edge Impulse Studio, which was used to generate spectral features from the three time-series sensor values. This information was then fed to a Keras neural network that had been configured to perform the classification, resulting in an overall accuracy of 92.86% when run on real-world test samples.
>
Confident in his now trained model, Nekhil deployed his model as an Arduino library on the Nano 33 BLE Sense. The Nano sends a message through its UART pins to a standby ESP8266-01 board when a fire has been detected. And in turn, the ESP8266 triggers an IFTTT webhook to alert the user via email.
If you want to know more about the construction of this fire recognition system, you will find many details on the project page.
What's Your Reaction?
