AI-Driven Early Fire Detection Using CubeSats: Enhancing Real-Time Satellite Monitoring for Wildfire Prevention

SmartSat is seeking applications for a hands-on project-based PhD looking at AI-Driven Early Fire Detection Using CubeSats: Enhancing Real-Time Satellite Monitoring for Wildfire Prevention.

This PhD will be facilitated through SmartSat and Uni SA STEM, in partnership with the European Space Agency (ESA) Phi-lab, Green Triangle Forest Industry Hub/Green Triangle Fire Alliance, South Australian Country Fire Service and the Victorian Country Fire Authority.

The successful applicant will undertake several significant stages of research, enhancing existing AI models and onboard processing systems for CubeSats, focusing on segmentation-based detection and integrating advanced deep learning techniques. The applicant will also help develop solutions that integrate thermal information, visible fire scars, fire smoke aerosols, and additional factors such as fuel and fire risk to achieve a comprehensive onboard wildfire detection system.

This project provides extensive opportunities to collaborate with academic and industry partners, engaging through workshops, live demonstrations, and conferences. Research may involve travel, particularly to ESA’s Phi-lab in Frascati, Italy for research cooperation and technical knowledge exchange.

Applicants must meet the eligibility criteria for entrance into a PhD. All applications that meet the eligibility and selection criteria will be considered for this project. 

This project is open to applications from both Domestic and International applicants, provided they meet the project selection criteria: 

• Educational background in Remote Sensing or Machine Learning
• Excellent programming skills

The successful applicant is expected to study full-time and to be based at our Mawson Lakes campus in the north of Adelaide. Applicants are expected to start in a timely fashion upon receipt of an offer. Extended deferral periods are not available. Applications close Sunday, 1 December 2024.  

For more information, contact Dr Stefan Peters, Senior Lecturer of Geospatial Science, at [email protected] or click here.

PHD CANDIDATES FROM THE UNIVERSITY OF SOUTH AUSTRALIA

Mohamad Abdul Hady

SCARLET-α: SpaceCraft Autonomy and Onboard AI for Next Generation Space Systems

University of South Australia

Franke Agenbag

An automated method of detecting, characterising, and responding to radiation events in space

Defence, Science and Technology Group / University of South Australia

Xiongren Chen

Interpretable Machine Learning for the Early Smoke Wild-fire Detection

University of South Australia

Joshua Davis

Attack-resilient CubeSat Constellations

University of South Australia

Donna Fitzgerald

Quantifying Dieback of Eucalypt Forests using Remote Sensing

University of South Australia

Monirul Islam Pavel

SCARLET-⍺: SpaceCraft Autonomy and Onboard AI for Next Generation Space Systems

University of South Australia

Uakomba Uhongora

Development of Intelligent Security Controls for Software Defined Networking-based Space Systems

University of South Australia

Liang Zhao

Satellite image-based smoke detection for bush fire detection

University of South Australia

Artur Medon

Small Satellite Thermal Management with 3D Printed Metal Heat Sinks Containing Phase Change Material Thermal Storage

University of South Australia

Jordan Plotnek

Measuring Control System Resilience to Cyber-Physical Threat in a Satellite Context

University of South Australia

Ahsan Waqas

Distributed Beamforming for Satellite Applications

University of South Australia

RELATED NEWS

Visualisation shows change and anomaly detection in observed ISR patterns

Research outcomes opening doors for future projects

Cross-Sensor Transfer Learning for Fire Smoke Scene Detection Using Variable-bands Multi-spectral Satellite Imagery