Google Earth images before and after the Mud Creek landslide that occurred in February 2017 near Big Sur, California. The Penn State team's effort will catalog and improve our ability to predict landslide events.
Penn State is a Google AI Impact Grantee
Penn State College of Engineering will apply AI to develop more accurate landslide alert system
05/07/2019
UNIVERSITY PARK, Pa. – Google announced today that Penn State is one of 20 organizations that will share $25 million in grants from Google.org, credit and consulting from Google Cloud and coaching by Google’s AI experts as a grantee of the Google AI Impact Challenge.
The Google AI Impact Challenge was an open call to nonprofits, social enterprises and research institutions from around the world to submit their ideas to use AI to help address societal challenges. More than 2,600 organizations applied.
Landslides, including mud flows and rock avalanches, can mobilize large volumes of geomaterials and are one of the most widespread hazards that face the world. Penn State will receive a $750,000 grant to train deep learning (DL) networks to build a global-scale database of landslide/mudslide occurrences using Google Earth Services. They will then train DL models to predict landslide location, timing, rainfall thresholds and impacted areas and provide uncertainty estimates. Lastly, working with Google and the U.S. Geological Survey, they hope to integrate their findings into a warning system.
"This research is so important because there is a good chance that we can advance the science and do something that really would impact the lives of many people,” said Chaopeng Shen, associate professor of civil engineering and principal investigator of the project. “In developed countries like the U.S. and many developing countries, we have had quite a few of these events that, unfortunately, kill many people."
Shen will join Penn State College of Engineering researchers Daniel Kifer, associate professor of computer science, and Tong Qiu, associate professor of civil engineering, on this investigation.
"We’re immensely proud of the Penn State team,” said Justin Schwartz, the Harold and Inge Marcus Dean of Engineering. “To be selected for the Google AI Impact Grant speaks highly not only of the potential of our engineers’ work to better understand these potentially tragic landslide/mudslide occurrences, but also of their ability to take what they learn to engineer solutions that will significantly impact society. It's an inspiration for all of us, and especially for their students."
Next week, Qiu and Shen will travel to San Francisco to dive into execution. For five days, all 20 organizations will join Google AI experts, project managers and the startup specialists from Google’s Launchpad Accelerator for a program that will last six months, from May to November 2019. Through the Launchpad program, each of the 20 grantees will develop their own objectives and key results and set timelines for project completion. Each organization will be paired with a Google expert who will meet with them regularly for coaching sessions and will also have access to other Google resources and expert mentorship.
“We received thousands of applications to the Google AI Impact Challenge and are excited that Penn State was selected to receive funding and expertise from Google,” said Jacquelline Fuller, president of Google.org. “AI is at a nascent stage when it comes to the value it can have for the social impact sector, and we look forward to seeing the outcomes of this work and considering where there is potential for use to do even more.”
Google.org, Google's philanthropy, supports nonprofits that address humanitarian issues worldwide and apply radical, data-driven innovation to solving the world's biggest challenges.
Share this story:
From left: Tong Qiu, Chaopeng Shen and Daniel Kifer, recipients of the Google AI Impact Grant.
