AI to fight forest fires – USC Viterbi


SoCal Fire, Photo credit: Eddiem360, CC BY-SA 4.0 , via Wikimedia Commons

Over the past decade in Los Angeles and the state of California, the question is not whether there will be wildfires, but rather when and where they will sprout and how to protect people from these threats. . As such, firefighters must know how to plan and deploy limited resources.

One of these solutions is to control flammable brush burns to avoid worst case scenarios of tinder growth left unattended, providing fodder for mega fires. With $ 5 million in support from the National Science Foundation’s Convergence Accelerator program, a team of researchers, which includes the San Diego Supercomputer Center (SDSC) at UC San Diego, the Viterbi School of Engineering at the University of Southern California, and the Tall Timbers Research Station in Florida, will bring the power of AI to help firefighters strategize to best plan these controlled fires, as well as manage unexpected fires.

SDSC will lead efforts through the development of “BurnPro3D”, a new decision support platform to help the fire response and mitigation community quickly and accurately understand the risks and tradeoffs presented by fire to more effectively plan controlled burns and manage forest fires.

The BurnPro3D platform will leverage SDSC’s WIFIRE Commons, a data sharing and AI framework that uses next-generation fire science in prescribed burns for preventative vegetation treatment and MINT’s modeling framework. USC, which integrates very heterogeneous models from distinct disciplines, including geosciences, agriculture, economics and social sciences.

Ilkay Altintas, Data Science Officer and Director of the WIFIRE Lab at SDSC, is the Principal Investigator (PI) of the project.

For the USC team, Yolanda Gil, director of new initiatives in AI and data science at USC’s Viterbi School of Engineering, will be the principal investigator.

“The NSF Convergence Accelerator program is focused on innovation for societal impact. We have been developing key infrastructure and partnerships in this area for several years, and more recently we have worked with collaborators from USC to include AI in various aspects of the project. said Altintas.

USC’s Gil, a leading figure in AI, has long used AI to solve problems related to natural resources and the environment. Gil explains that AI can be used to perform automated reasoning on factors such as wind speed and direction, slope, and vegetation type and density, so that he can quickly create accurate models of the wind. evolution of a controlled fire under different initial conditions. Additionally, says Gil, AI will allow decision-makers to customize their mitigation strategies, for example, by creating a custom plan if the preference is to burn only 20% of the vegetation in one location or reduce the impact on it. air quality.

The project also has personal meaning for Gil, a longtime resident of the Los Angeles area who has seen the impact of natural disasters in California but also in her native Spain, which was recently hit by volcanic eruptions on the islands. Canary Islands. Her first work on AI for natural disasters dates back to the mid-1980s as a student intern, where she worked with a civil engineering research group on the use of expert systems to predict when flooding occurs. frequently in the Ebro, the longest river in Spain.

In addition to Gil, USC brings major AI expertise to the direction of the project. Bistra Dilkina, co-director of the Center for AI in the Society, will contribute to this effort with Michael Pazzani of the USC Information Sciences Institute. USC will contribute to AI research in three key areas:

• use automated reasoning to select the best fire models for the conditions in a given area and access the necessary data;
• integrate physics-based machine learning with next-generation fire models and deep learning to understand the complex processes that determine fire behavior;
• apply stress optimization methods to resolve complex trade-offs in the decision-making process for the placement and timing of controlled burns

Like Gil, Dilkina has also long applied AI to understand the impacts of natural disasters and has also used AI to respond to wildlife conservation efforts. His efforts will focus on the optimization part of this work.

Dilkina says: “Selecting the most efficient spatial distribution of prescribed burns is a difficult task, as it involves complex fire dynamics, limited resources and multiple objectives. I’m excited to help deliver AI approaches to provide much needed decision support to agencies responsible for wildfire mitigation and response. “

Pazzani, a leading machine learning researcher, brings to the project deep expertise in transparency and explanation, which are crucial for supporting AI decision support systems.

BurnPro3D is one of 10 multidisciplinary research proposals that recently received collective support of $ 50 million from NSF, funded by the Convergence Accelerator program. According to Douglas Maughan, head of the NSF program, a convergence approach is essential to solving large-scale societal challenges, such as the massive fires that regularly burn in the western United States.

Over the next two years, these teams will participate in an innovation and entrepreneurship program that includes product development, intellectual property, financial resources, sustainability planning, as well as communications and outreach.

“The fusion of ideas, techniques and approaches combined with human-centric design concepts and end-user insight helps our teams turn their ideas into proof of concept, then prototype and finally in solution. With a three-year research model, we expect these teams to deliver high impact deliverables, ”said Maughan.

The NSF launched the first phase in 2020 when 29 teams received support to develop their concepts and build their groups.

Source link

Leave A Reply

Your email address will not be published.