Researchers at Virginia Tech developed artificial intelligence tools that measure surface vibrations to determine how deeply rocks are buried beneath planetary terrain. Yiyan Ruan, a Ph.D. candidate, and Erik Komendera, an assistant professor of mechanical engineering, published their findings in IEEE Sensors Journal.
The team buried 18 concrete blocks of different shapes and sizes in sand at varying depths. They struck the exposed portions with an instrumented hammer and measured vibrations using a Laser Doppler Vibrometer. A neural network called a Multilayer Perceptron processed 11 measurable features to map the relationship between vibration patterns, rock geometry, and burial depth.
"Estimating how deep a rock is buried using vibrations can be tricky," Ruan said. "The feedback is affected by the rock's geometry and the way the soil holds it. By carefully selecting features from vibration measurements and exposed geometry, we were able to produce reliable data using a compact neural network with just two hidden layers."
The models could help future rovers on Mars determine obstacle depth and plan excavation autonomously, avoiding energy waste on rocks too deeply buried to move. Komendera noted the challenge: "It's easy enough for us to just look at rocks and make some quick, testable judgments. Is this rock resting or partially buried? Can I move this alone or do I need help and tools? But what if robots need to do this? How can robots make those judgments? Now we have an answer."
The team plans to expand their datasets and test applications on Earth, including analyzing construction sites, natural disasters, unstable rock formations, and underground caverns.