Carnegie Mellon University

photo of road destroyed by landslide

Using Machine Learning to Prevent Landslides

Landslides in the United States are becoming increasingly more common and more severe due to climate change. The United States Geological Survey estimates that each year between 25 and 50 deaths are due to landslides, as well as between $2 billion and $4 billion in annual losses due to property damage. 

In 2018, Pennsylvania’s Allegheny County experienced an unprecedented number of landslides, resulting in damage to at least 131 properties with an estimated cost to fix the county’s landslide-related damage of about $40 million. This amount may seem daunting, however, according to Karen Lightman, executive director of Metro21: Smart Cities Institute, 2018 wasn’t an outlier — it’s the new normal.

Christoph Mertz, the principal project scientist at Carnegie Mellon University’s Robotics Institute, started taking pictures of the hills overlooking Pittsburgh’s West End on his smartphone. Mertz’s curiosity eventually turned into a research project sponsored by Metro21. “Every day, for months, I was collecting images of these hillsides,” Mertz said. “I wanted to see if I could use these pictures as a way to predict the next landslide.”

Mertz is no stranger to finding innovative ways to anticipate infrastructural decay. In addition to his role at the Robotics Institute, Mertz is the co-founder of RoadBotics, previously funded by Metro21, where he uses deep learning analysis of smartphone images to identify developing potholes and other road infrastructure issues in real time. More than 100 governments around the world now use RoadBotics’ pavement assessment system.

Mertz wondered if he could use the same deep learning approach used for RoadBotics to detect signs of impending landslides, like fast developing cracks in the road, deformed guard rails, debris on the road, deformation of hillsides or tilting trees.

The project is now underway through the support of Metro21. Mertz said “You need the kind of cross-discipline collaboration that’s here at Carnegie Mellon University — not just experts in computer science and machine learning but experts in geology, in infrastructure, in water and sewage — to come together and tackle the issue.”

Ultimately, Mertz’s project is not only about being able to predict and prevent landslides. He also intends to use this work to more equitably direct the infrastructural change necessary to support this kind of prediction and prevention.

“I’m not sure that landslide prevention was in the vernacular even three years ago,” Lightman said. “But now, I’m hearing it more often in conversations about future investments in the infrastructure.”

For more information click here