X-Ray Vision Robots Can See Through Walls Using Only WiFi

University of California Santa Barbara researchers have demonstrated a system for two unmanned vehicles to see through thick concrete walls using only WiFi signals - fully discovering what is on the other side with high accuracy. The objects on the other side don't have to move to be detected. Potential applications include search and rescue operations and surveillance. One robot measures the wireless transmissions of the other robot that is in the broadcast mode. Each wireless transmission passes through the unknown area and the objects attenuate the signal depending on their material properties and locations. By devising a framework based on proper wave propagation modeling and sparse signal processing, the researchers have shown that it is possible for the two unmanned vehicles to image the entire area, with a high targeted resolution, and see through highly-attenuating walls. They used directional antennas in order to increase the imaging resolution.

Transcript

00:00:00 [Music] me us to a man Vehicles they're outside of an area that is completely unknown to them yet they're interested in fully Imaging it at ctim level solution but they only have Wi-Fi cars for imag and no other sensors the area can also have several uded parts are parts that are entirely behind walls and invisible to humans

00:00:30 robots or any Noe outside so this is a very fundamental problem of extracting information about our environment from Wireless transmissions and my group has been working on it for the past few years so let's say the two robots move outside of the area of Interest while one transmits a Wi-Fi signal and the other one receives it that transmission

00:00:55 goes through the objects in the area of interest each object leads a signature on the transmitted RF signal by interacting with it depending on its material property or location but this information is very implicit and mixed up in the receive signal also they're lot more unknowns than the number of RF measurements that these robots can collect so we have proposed a framework

00:01:21 based on proper approximated wave models to capture what goes through the structure Spar signal processing to address the huge size of the unnown space and path planning and we have shown that it's actually possible to do this path for instance here the robots are traveling for Round We can see snapshots of their

00:01:44 Imaging results as they collect more Wi-Fi measurements the robots do not coordinate anything while traveling each round each robot mathematic L estimates its own position and where the other robot would be at the end they collect less than 2% Wi-Fi measurements which means that we have a lot more unknowns than the number

00:02:14 of collected measurements yet we can see that high resolution imaging is achieved we have tested our approach with several different structures and material here's a more complex and bigger structure they can image that too let's say there's a human [Music] inside here's the high resolution

00:02:52 imaging result the applications are very exciting and unimaginable um from robotic search and rescue to let's say a smart um Health monitoring system let's say that uh for an elderly monitoring you use the existing Wi-Fi network to attract the person it's a really really exciting time to work in this area because there's so many unanswered

00:03:18 questions and at UCSB we have a great team of passionate students that are working to solve these problems [Music]