X-ray vision will let soldiers see through walls

Soldiers and police could gain superhero-style X-ray vision thanks to a system that uses radio waves to map enemies’ movements behind walls.

Scientists have used artificial intelligence to teach a wireless device to detect people’s precise actions and gestures even when they were behind cover.

The team at Massachusetts Institute of Technology (MIT) used a ‘‘neural network’’ to analyse radio signals 1,000 times less powerful than home wifi, which were bounced off bodies to create animated stick figures that walk, sit and move their limbs in sync with the targets.

The system can also identify individuals based on their size and characteristic movements.

In future, a soldier or police marksman wearing a transmitter device could ‘‘see’’ projections of concealed enemies in their field of vision by wearing augmentedreality goggles. With further training, the system could create more detailed, 3D figures.

Previous versions of the technology could reveal the position of a person behind an obstacle, but not their precise movements.

The MIT team trained the AI to translate radio signals bounced off humans into the corresponding movement of 14 key points on the body, including the head, elbows and knees.

Dina Katabi, the lead researcher, told technology website Motherboard: ‘‘Around 2013, we showed that we can track people accurately. What’s new is that, for the first time, we can create a dynamic skeleton of the person, their posture, and how they’re moving.’’

This made the new technology far more useful, she said.

‘‘Let’s say the police want to use such a device to see behind a wall. It’s very important to know if somebody is standing in a position that indicates they are holding a weapon, for example. All of that you can’t do with just a blob.’’

One challenge the researchers had to address was that most neural networks – a form of artificial intelligence based on the human brain – are trained using data labelled by hand. A network trained to identify cats requires that people look at a big dataset and label each image as ‘‘cat’’ or ‘‘not cat’’.

Radio signals aren’t easily labelled by humans. To address this, the researchers collected data using both their wireless device and a camera.

They gathered thousands of images of people walking, talking, sitting, opening doors and so on. This enabled the AI to learn the association between radio signals and the movements of individuals, represented as stick figures.

After the training, the technology was able to estimate a person’s posture and movements without cameras, and behind walls, using only the wireless reflections that bounced off people’s bodies. In addition to tracking precise movements, it could accurately identify individuals based on their size and gait.

Although the researchers have cited policing applications, their paper does not discuss military usage and focuses on healthcare.

The team says the technology could be used in the home to monitor diseases like Parkinson’s and multiple sclerosis, providing a better understanding of disease progression and allowing doctors to adjust medications accordingly. It could also help frail people to live more independently, while providing monitoring for falls, injuries and changes in activity patterns.