A team at the Massachusetts Institute of Technology has created a new form of “X-ray vision” that could dramatically change how robots perceive the world around them. Using a system called mmNorm, researchers have found a way to reconstruct the shape and orientation of objects hidden behind walls or inside containers using wireless signals similar to those used in Wi-Fi and 5G. This breakthrough allows robots to generate detailed 3D models of concealed objects without ever laying eyes on them.
The technology relies on millimeter-wave signals, which can pass through non-metal materials like cardboard, plastic, and standard interior walls. Traditional radar systems are capable of detecting the presence and general location of hidden objects, but they struggle to produce more than rough outlines. MIT’s innovation lies in how the mmNorm system analyzes signal reflections, treating surfaces like mirrors for wireless waves. By capturing how these signals bounce off various surfaces from multiple angles, the system can calculate both the orientation and curvature of each object point. These calculations are then used to reconstruct a full 3D image of what lies beyond the surface.
In laboratory tests, mmNorm demonstrated a stunning 96% accuracy in reconstructing a wide variety of objects—from silverware to complex power tools—far surpassing the precision of previous wireless imaging systems. This capability opens up a host of real-world applications, especially in robotics and automation. Imagine a robot in a warehouse being able to identify not just that an item is inside a sealed box, but exactly what the item is, whether it’s facing the right direction, or if it’s been damaged—all without opening the package.
The potential goes well beyond robotics. The MIT team envisions augmented reality headsets for factory workers that could overlay real-time visuals of machine components hidden inside large equipment, making diagnostics and repairs more efficient. In airport security, improved scanners using this technology could provide clearer images of concealed items without the need for invasive searches. While the system currently can’t see through metal or very thick walls, researchers are actively working on extending its capabilities.
Lead author Laura Dodds described the project as a paradigm shift in how wireless signals can be used. “This work really represents a paradigm shift in the way we are thinking about these signals and this 3D reconstruction process,” she said. The research team believes that the insights gained from mmNorm could have a broad impact across several industries, from manufacturing and logistics to defense and medical diagnostics.
MIT’s mmNorm system is still in development, but its results are already pointing to a future where robots and smart systems can gain an almost superhuman level of environmental awareness—powered not by sight, but by the invisible waves all around us.