Revolutionizing Satellite Real-Time Air Pollution Alerts through New On-Orbit System-on-Chip Technology

Exposure to abnormally high concentrations of particulate matter and ozone can cause severe harm to human health, highlighting the need for real-time satellite monitoring to enable rapid responses and timely warnings. However, the existing methods for on-orbit diagnostics under resource constraints are limited. This study presents SoC-POM (System-on-Chip for Particulate Matter and Ozone Monitoring), a real-time, on-orbit artificial intelligence algorithm embedded in satellites and designed to detect anomalous concentrations of PM_2.5, PM₁₀, and O₃. Based on tests with the Himawari-8 and Himawari-9 satellites, we demonstrate that SoC-POM achieves an average latency of 5.5 min, breaking through the hourly processing barrier while maintaining high accuracy, with correlation coefficients of 0.78, 0.76, and 0.81 for PM_2.5, PM₁₀, and O₃, respectively. This novel approach enables real-time monitoring of abnormal particulate matter and ozone levels and demonstrates the potential for the timely analysis of health exposure and its dynamic changes, marking a sustainable advancement in air pollution alerts and public health.