NVIDIA & UR join forces with Inbolt for intelligent automation
Bridging the gap between humans and robots: Inbolt, Universal Robots, and NVIDIA bring vision to factory floors.
Manufacturers have long struggled to automate tasks that demand flexibility and perception, where parts move, vary, or change over time. Traditional robots simply couldn’t keep up. But thanks to Universal Robots’ AI Accelerator, powered by NVIDIA Jetson and Isaac technologies, and Inbolt’s 3D vision AI, that’s changing fast.
By combining Inbolt’s vision-based intelligence with NVIDIA’s edge AI performance, robots can now see, think, and adapt almost like humans. This unlocks automation for complex processes once considered impossible, from dynamic part handling to precise assembly on moving lines.
“Before integrating NVIDIA Jetson, our depth image computation took 100 milliseconds. Now it’s just a few,” explains the Inbolt’s CEO Rudy Cohen. “That leap in performance means robots can achieve the reliability required in real manufacturing environments, up to 99.99%.”
This new generation of intelligent robot systems, built on Universal Robots’ AI Accelerator platform, demonstrates how close collaboration between industry leaders can transform what’s possible in automation.
Inbolt’s partnership with Universal Robots and NVIDIA isn’t just about faster computing. It’s about empowering manufacturers to deploy automation that adapts to the real world without costly reprogramming or downtime.
The result is robots that no longer just repeat tasks but truly respond to their environment.
Watch the video below for the full story.
Explore more from Inbolt
Access similar articles, use cases, and resources to see how Inbolt drives intelligent automation worldwide.
The Stellantis story the New York Times told, and the engineering behind it
The gap between a great demo and a robot that runs in production
Oltre 106.000 robot lavorano già nelle fabbriche italiane. Quasi tutti alla cieca.
Inbolt Robot Programming: from digital twin to a running robot in five minutes
The missing execution layer
Why real-time robot control is the key to unlocking Physical AI
What bin picking really means in production: from semi-structured to unstructured environments
