Article image

Bushfires are detected 500 times faster with AI satellites

A significant advancement is underway in the detection of bushfires, as Australian researchers develop cube satellites equipped with AI that can identify bushfires from space at unprecedented speeds.

These satellites, leveraging advanced technology, can detect fires 500 times faster than the traditional methods of processing imagery on the ground. Rapid detection is critical in managing and mitigating the devastating impacts of wildfires.

Bushfires and their impacts

Bushfires, also known as wildfires, are uncontrolled fires that burn through vegetation, forests, grasslands, or other landscapes. They can start from natural causes like lightning or human activities such as campfires, discarded cigarettes, or arson.

Bushfires spread rapidly, especially in hot, dry, and windy conditions. They can cause widespread destruction to ecosystems, wildlife, property, and human lives. The intense heat can create its own weather patterns, making the fires unpredictable and difficult to control.

Effective management includes early detection, swift response, and preventive measures like controlled burns and public education to reduce the risk and impact of bushfires.

Cube satellites and bushfires

The key to an accelerated bushfire response lies in the innovative use of smaller, cost-effective cube satellites.

The satellites, despite their size, are equipped with the capability to process and compress large volumes of hyperspectral imagery directly on board.

This method circumvents the previous limitations where data had to be sent to Earth for analysis, thus saving crucial time and resources.

Early bushfire detection with AI satellites

The breakthrough comes from the implementation of AI directly within the satellite’s systems. This allows for the detection of bushfire smoke even before the fire escalates and becomes more visible and heat-intensive.

The early detection enables ground crews to respond swiftly, potentially saving lives and preventing extensive property damage.

This project is a collaboration among the South Australian Government, SmartSat CRC, and various industry partners. Together, they have launched the Kanyini mission, which is South Australia’s first cube satellite.

Named Kanyini, this 6U CubeSat is tasked not only with bushfire detection but also with monitoring inland and coastal water quality.

Innovations in satellite imaging

The hyperspectral imager on the satellite plays a crucial role. It captures light reflected from Earth across various wavelengths, producing detailed maps that are invaluable for tasks like monitoring water quality, land management, and detecting bushfires.

This satellite marks a significant step forward in the capability of Earth observation technologies.

The research team, led by Dr. Stefan Peters from the University of South Australia, includes experts from Swinburne University of Technology and Geoscience Australia.

The experts have developed a lightweight AI model tailored to the processing capabilities and energy constraints of cube satellites. This model significantly reduces the volume of data that needs to be transmitted back to Earth and cuts energy consumption by nearly 70%.

Enhancing bushfire detection with satellites

The onboard AI model is not just faster but also more data-efficient. It has successfully demonstrated the ability to reduce the data sent to Earth to just 16% of its original volume – all while maintaining the speed and accuracy needed for effective fire detection.

This was vividly demonstrated using simulated imagery from recent Australian bushfires, where the AI was trained to distinguish smoke from clouds, enhancing both speed and efficiency of detection.

A practical demonstration of this technology used a past bushfire event in the Coorong region as a case study.

The Kanyini satellite managed to detect smoke and transmit the data back to a ground station at the South Pole in under 14 minutes—a testament to the system’s efficacy.

AI-driven future

Looking ahead, the research team plans to fully operationalize the onboard AI fire detection system with the Kanyini mission by 2025.

The experts aim to commercialize the technology and potentially employ it in a constellation of CubeSats dedicated to early fire detection. The goal is to detect fires within an hour of their start.

The implications of this technology extend beyond bushfires. It holds promise as an early warning system for a variety of natural disasters, offering a new paradigm in disaster management and response.

The study is published in the IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.


Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates. 

Check us out on EarthSnap, a free app brought to you by Eric Ralls and


News coming your way
The biggest news about our planet delivered to you each day