Remote Sensing for Forest Fires
In the midst of global warming, British Columbia and Canada have been experiencing a troubling rise in forest fires. The destruction of wildfires impacts ecosystems and communities. In the effort to combat the threat of increasing fires, the UBC CIC submitted this real-world problem as a challenge for the Department of Electrical and Computer Engineering capstone students to apply their skills and solve for, under guidance of their instructor and the CIC team. The CG-23 team developed FireWatch, offering early prediction and detection of fires. The prototype allows rapid firefighting responses to protect and bring awareness for at-risk communities and smoke hazards.
— Capstone Team CG-23
Remote Sensing for Forest Fires
Published:
The video above was created by capstone team CG-23. The video describes the goal of the project and includes a demonstration of the open-source solution.
Often, traditional wi-fi and cellular networks are unstable in vast, uninhabited lands prone to wildfires. The capstone team CG-23 developed the application, FireWatch, to overcome the challenge by including satellite data and using low-power and wide-area network protocols to ensure reliable wildfire monitoring in isolated regions. The Remote Sensing for Forest Fires system predicts and detects wildfires in British Columbia and Canada. The deployable prototype uses IoT sensors to remotely collect near real-time forest and satellite data to monitor fires. The data collected is displayed on the web application for users and notifications are sent to inform them of high-risk areas.
In addition to enabling stable surveillance and early detection, the solution is designed for scalability and is open-source. The features promise better protection for at-risk communities and encourage collaborative development in wildfire prevention and control.
Approach
The UBC CIC sought to contribute to the reduction of wildfires, which are an increasing concern, and boost community safety and wellbeing. The challenge was shared by the UBC Department of Electrical and Computer Engineering.
Solution Overview
The solution consists of data collection, data transmission, data storage and processing, and data display.
Screenshots of UI
Architecture Diagram
Technical Details
Expand below to learn more about the architecture diagram, step-by-step.
Data Collection
The system focuses on detecting and predicting through a strategic integration of IoT sensors and satellite imaging. The BME688 IoT sensor is employed to collect environmental data such as temperature, humidity, and pressure. The sensors are important for near real-time monitoring, which allows for early wildfire detection and facilitation of response actions. To address the limitations of the sensors (i.e., comprehensive environmental assessments), satellite data is obtained through the OpenWeather API. The satellite data includes the Fire Weather Index (FWI), which plays a significant role in wildfire risk evaluation.
Data Transmission
Long Range
Long Range Wide Area Network (LoRaWAN) is used so that the data can be transmitted to the web server from a user remotely monitoring the scene, devoid of traditional wi-fi coverage. The SenseCAP M2 Multi-Platform LoRaWAN Indoor Gateway allows the system to function over long distances usually found in fire-prone areas. The gateway has appropriate range and can connect multiple devices, making the application scalable. The EU868 ISM band enables the gateway to gain reliable and better range and network penetration capabilities, which are necessary for dynamic terrains where wildfires may occur.
Microcontroller
The microcontroller is on the sensor device, and is responsible for the transmission of the data to the gateway. The Wio-E5 mini Dev board allows the function of the gateway capability and is built around the STM32 microcontroller. The board is cost-effective, compact, and consumes minimal power. The microcontroller allows the integration of sensor data to be reliably communicated to the gateway, then onto the web for real-time monitoring and analysis.
Data Storage and Processing
AWS
The services are the foundation for data storage and processing, ensuring a scalable, secure, and reliable infrastructure. The team uses Lambda functions to integrate data ingestion from the OpenWeather API and the sensor network. The functions ingest the data and transforms it into a structured format appropriate for analysis and storage into DynamoDB. Overall, the setup allows for high availability and quick data retrieval.
Machine Learning Model
The model uses a Random Forest classifier and is trained to accurately discriminate between high- and low-risk scenarios. The model continuously learns from incoming data, which in turn enhances its predictive accuracy overtime.
Data Display
The system features a web application that presents sensor and satellite data to end-users, developed using ReactJS. The data stored in DynamoDB is retrieved through Lambdas, securely transmitting the data to the front-end through the API Gateway. The team also employed the Simple Notification Service (SNS) to alert users right when the machine learning model identifies a high risk of fire. Through this, users can take immediate action based on real-time data and predictions of wildfires.
Link to solution on GitHub: https://github.com/UBC-CIC/Forest-Fire-Sensing
Acknowledgements
This application was developed by senior students in Electrical and Computer Engineering as part of the UBC Electrical and Computer Engineering Capstone Program. Guidance was provided by a faculty member who acted as the technical director and supported by the UBC Cloud Innovation Center technical team.
Photo by Andrii Chagovets.
About the University of British Columbia Cloud Innovation Centre (UBC CIC)
The UBC CIC is a public-private collaboration between UBC and Amazon Web Services (AWS). A CIC identifies digital transformation challenges, the problems or opportunities that matter to the community, and provides subject matter expertise and CIC leadership.
Using Amazon’s innovation methodology, dedicated UBC and AWS CIC staff work with students, staff and faculty, as well as community, government or not-for-profit organizations to define challenges, to engage with subject matter experts, to identify a solution, and to build a Proof of Concept (PoC). Through co-op and work-integrated learning, students also have an opportunity to learn new skills which they will later be able to apply in the workforce.
