Are you looking for a holiday? Get special deals.

New technology helps robots see inside sealed boxes

14 Jul 2025 By foxnews

Why would you want a robot to see inside a sealed box?

Digging deeper beyond the initial creepy thought, a new breakthrough from MIT could soon let warehouse robots do something pretty remarkable. This high-tech wave technology can detect damage inside sealed cardboard boxes without ever opening them.

Using millimeter wave (mmWave) imaging, the same type of signal used in Wi-Fi, the researchers developed a system called mmNorm. There's nothing normal about this superpower. This technology allows robots to scan containers and generate accurate 3D models of the objects inside by analyzing how signals bounce back from hidden surfaces.

Sign up for my FREE CyberGuy Report
Get my best tech tips, urgent security alerts and exclusive deals delivered straight to your inbox. Plus, you'll get instant access to my Ultimate Scam Survival Guide - free when you join my CYBERGUY.COM/NEWSLETTER

NONINVASIVE BRAIN TECH AND AI MOVES ROBOTIC HAND WITH THOUGHT

Millimeter waves are particularly effective at penetrating materials such as cardboard, plastic and even interior walls. When these waves hit an object inside a box, they reflect. The mmNorm system captures these reflections and feeds them into an algorithm that estimates the shape and direction of the surface of the hidden item.

Unlike traditional radar systems, mmNorm accounts for a challenging property called specularity, which refers to the way radio waves bounce off shiny or angled surfaces, much like a mirror. By estimating the surface normal (the direction a surface is facing), the system's accuracy improves dramatically. The method combines reflections from multiple antennas that "vote" on the surface normal direction based on signal strength, improving the accuracy of the 3D reconstruction.

WHAT IS ARTIFICIAL INTELLIGENCE (AI)?

What makes mmNorm different is its clever use of physics. Instead of ignoring reflections that bounce away from the radar (as most systems do), it captures them to infer the shape and orientation of hidden surfaces. Lead author Laura Dodds explained that the system doesn't just track where a signal comes from. It also analyzes the direction the surface is facing. To do this, the MIT team used a robotic arm equipped with a radar unit. As it moved around a sealed box, it gathered multiple measurements, building a detailed 3D picture of what was inside.

During testing, mmNorm achieved 96% accuracy in reconstructing complex items, such as power drills and silverware. That's a big jump compared to the 78% accuracy offered by similar systems. In a warehouse setting, this means robotic arms could inspect the contents of packages moving down a conveyor belt without unpacking them.

It also opens the door for humanoid robots to roam warehouses and assess packages for damage or missing parts in real time. However, the technology does not work well for objects hidden behind metal or very thick walls, which limits some applications.

GET FOX BUSINESS ON THE GO BY CLICKING HERE

Beyond warehouse automation, this technology opens the door to several other promising uses. In factory production lines, robots that detect damage inside boxes could inspect goods quickly and without opening packages, helping maintain high-quality standards.

In assisted-living centers, the same approach might help ensure safety by checking the contents of containers without disturbing residents. Even in security screening, being able to see through sealed boxes could enhance threat detection without adding any strain on bandwidth since the system uses the same mmWave signals as existing technologies. The research team plans to further improve the resolution of mmNorm and boost its performance on less reflective objects, making it even more versatile for future applications.

We may be entering an era where robots don't just move boxes, but they understand what's in them. MIT's mmNorm system offers a glimpse into the future of warehouse efficiency, safety and automation. And if robots can detect damage inside boxes without opening them, that might just change how everything gets shipped, sorted and delivered.

Are we ready to trust machines to judge what's damaged and what's not without ever opening the box? Let us know by writing us at Cyberguy.com/Contact