Table of Contents
Fetching ...

Hierarchical Framework for Optimizing Wildfire Surveillance and Suppression using Human-Autonomous Teaming

Mahdi Al-Husseini, Kyle Wray, Mykel Kochenderfer

TL;DR

A hierarchical framework outperforms firefighting techniques and a myopic baseline by up to 242% for moderate wildfires and 60% for rapid wildfires when simulated in abstracted and actual case studies.

Abstract

The integration of manned and unmanned aircraft can help improve wildfire response. Wildfire containment failures occur when resources available to first responders, who execute the initial stages of wildfire management referred to as the initial attack, are ineffective or insufficient. Initial attack surveillance and suppression models have linked action spaces and objectives, making their optimization computationally challenging. The initial attack may be formulated as a multi-agent partially observable Markov decision process (MPOMDP). We divide the initial attack MPOMDP into surveillance and suppression processes with their respective planners operating on different, but constant, time scales. A hierarchical framework iterates between surveillance and suppression planners while also providing collision avoidance. This framework is exemplified by a set of multi-rotor unmanned aircraft surveying an initial attack fire while a manned helicopter suppresses the fire with a water bucket. Wildfire-specific solver extensions are formulated to reduce the otherwise vast action spaces. The hierarchical framework outperforms firefighting techniques and a myopic baseline by up to 242% for moderate wildfires and 60% for rapid wildfires when simulated in abstracted and actual case studies. We also validate the early dispatching of additional suppression assets using regression models to ensure wildfire containment to thresholds established by wildfire agencies.

Hierarchical Framework for Optimizing Wildfire Surveillance and Suppression using Human-Autonomous Teaming

TL;DR

A hierarchical framework outperforms firefighting techniques and a myopic baseline by up to 242% for moderate wildfires and 60% for rapid wildfires when simulated in abstracted and actual case studies.

Abstract

The integration of manned and unmanned aircraft can help improve wildfire response. Wildfire containment failures occur when resources available to first responders, who execute the initial stages of wildfire management referred to as the initial attack, are ineffective or insufficient. Initial attack surveillance and suppression models have linked action spaces and objectives, making their optimization computationally challenging. The initial attack may be formulated as a multi-agent partially observable Markov decision process (MPOMDP). We divide the initial attack MPOMDP into surveillance and suppression processes with their respective planners operating on different, but constant, time scales. A hierarchical framework iterates between surveillance and suppression planners while also providing collision avoidance. This framework is exemplified by a set of multi-rotor unmanned aircraft surveying an initial attack fire while a manned helicopter suppresses the fire with a water bucket. Wildfire-specific solver extensions are formulated to reduce the otherwise vast action spaces. The hierarchical framework outperforms firefighting techniques and a myopic baseline by up to 242% for moderate wildfires and 60% for rapid wildfires when simulated in abstracted and actual case studies. We also validate the early dispatching of additional suppression assets using regression models to ensure wildfire containment to thresholds established by wildfire agencies.

Paper Structure

This paper contains 26 sections, 13 equations, 22 figures, 4 tables, 5 algorithms.

Figures (22)

  • Figure 1: Manned and unmanned aircraft coordinating to survey and suppress an initial attack fire.
  • Figure 2: The initial attack fire propagates based on fuel, winds, and terrain. Included in the initial attack environment are aerial agents surveying and suppressing the wildfire, and the resources that are subject to destruction.
  • Figure 3: The initial attack timeline depicts the arrival time for unmanned and manned aircraft, the escaped fire transformation, and the calculation and execution of surveillance and suppression actions.
  • Figure 4: Five suppression action drop-types and their associated aircraft axis of advance. A line drop and point drop have fundamentally different on-ground suppression characteristics.
  • Figure 5: Fuel, belief, and actual wildfire maps in two and three-dimensions as the initial attack fire propagates.
  • ...and 17 more figures