This project integrates advanced AI with geospatial analytics for real-time, high-precision natural disaster sensing. It leverages three AI models, with the aim of enhancing system robustness, maximising state estimation accuracy in sensor-scarce environments, and optimising sensor deployment.

The first model detects sensor anomalies, the second interprets geographic data for precise state estimations, and the third innovatively predicts the maximum mean-square error (MSE) to evaluate sensor deployment. This evolutionary algorithm-based optimisation framework tackles large-scale sensor network management, significantly improving disaster preparedness and responsiveness in New South Wales.