Robotic vision sensor camera system. (representational image) kynny/iStock
Robots still struggle with materials that the human eye can easily detect, such as glass, shiny metal, or jet-black surfaces.
These materials don’t reflect light in predictable ways, making them hard to read with standard 3D sensors. For industries that rely on speed and precision, that’s a serious problem.
Researchers at the Fraunhofer Institute for Applied Optics and Precision Engineering IOF have developed a breakthrough named ‘goROBOT3D’
This system cuts scan and evaluation time for uncooperative materials from 15 seconds to under 1.5 seconds.
The advancement could transform how automated systems interact with complex materials in dynamic production environments.
Making the invisible visible
Conventional systems fail on transparent or dark objects because they distort or absorb light. goROBOT3D gets around this by using structured thermal imaging.
It projects heat patterns onto a surface and captures them with thermal cameras.
The result is a clean 3D image of the object, regardless of how it looks to traditional sensors.
This thermal method has been in development for years. The latest leap comes from switching to a new kind of projection and image capture.
Single shot to capture 3D
The system now uses single-shot imaging. “With our method, the surface of the measurement scene is heated in a structured manner,” said Dr. Martin Landmann of Fraunhofer IOF.
“A statical thermal point pattern is emitted from the surface of the objects and recorded using two thermal imaging cameras. Using spatial cross-correlation, a 3D result can be obtained from the recorded image pair.”
The team replaced slow fringe projection with a fast, irregular dot pattern created by diffractive optical elements. These elements break a laser beam into a complex layout that hits the surface instantly.
This allows accurate measurement even in demanding industrial environments where speed and resilience matter.
The goROBOT3D uses intelligent thermal imaging to quickly capture and process hard-to-scan materials. Credit – Fraunhofer IOF
From scan to grip in milliseconds
Previously, scanning required hundreds of image pairs. Now it takes just one.
“Instead of capturing several hundred pairs of thermal images, as was the case with the previous method, our newly developed method can reconstruct the 3D information with just a single pair of images within a few milliseconds. This reduces the overall measurement and evaluation time by an order of magnitude,” Landmann explained.
RECOMMENDED ARTICLES
Once scanned, AI analyzes the object and selects gripping points. “We use a ‘bin-picking’ process for this,” he added.
“In other words, the targeted gripping of objects from chaotic environments.”
Ready for the factory floor
This speed opens the door to continuous production. Robots don’t have to wait. While one object is being picked, the next can already be scanned. That keeps operations moving without interruption.
Thanks to its modular design, goROBOT3D can be adapted for different industrial uses.
Fraunhofer IOF will demonstrate the tech live at automatica 2025 in Munich, from June 24 to 27.
MasterCard
