Automating the Wildfire Detection and Scheduling Pipeline with Maneuverable Earth Observation Satellites

TL;DR

The paper tackles the need for rapid, autonomous wildfire detection and timely data collection by integrating CNN-based wildfire detection, Bayesian multi-pass confidence updating, and a reconfigurable satellite-scheduling framework. The proposed WildFIRE-DS combines a Detection Module (YOLOv11 with sensor fusion, using Early or Late Fusion), a Multi-pass Confidence Module (Bayesian updating across flyovers), and a Schedule Module (REOSSP) to optimize orbital maneuvers, observations, and data downlinks. Experimental results show that Early Fusion with REOSSP yields the highest true-positive detections and data yield, significantly outperforming a baseline EOSSP and demonstrating the value of constellation reconfigurability and autonomous pipeline operation. The work offers a practical, autonomous workflow for near-real-time wildfire monitoring with potential to improve disaster response and environmental monitoring through onboard decision-making and optimized scheduling.

Abstract

Wildfires are becoming increasingly frequent, with potentially devastating consequences, including loss of life, infrastructure destruction, and severe environmental damage. Low Earth orbit satellites equipped with onboard sensors can capture critical imagery of active wildfires and enable real-time detection through machine learning algorithms applied to the acquired data. This paper presents a framework that automates the complete wildfire detection and scheduling pipeline, integrating three key components: wildfire detection in satellite imagery, statistical updating that incorporates data from repeated flyovers, and multi-satellite scheduling optimization. The framework enables wildfire detection using convolutional neural networks with sensor fusion techniques, the incorporation of subsequent flyover information using Bayesian statistics, and satellite scheduling through the state-of-the-art Reconfigurable Earth Observation Satellite Scheduling Problem. Experiments conducted using real-world wildfire events and operational Earth observation satellites demonstrate that this autonomous detection and scheduling approach effectively enhances wildfire monitoring capabilities.

Figures (13)

• Figure 1: A demonstration of the capabilities within the WildFIRE-DS.
• Figure 2: A simplified architecture flow diagram of YOLOv11.
• Figure 3: CNN training metrics over $100$ Epochs.
• Figure 4: CNN model responses to $1000$ validation images.
• Figure 5: Experiment orbital slot options and active wildfire parameters.