Universal Robots needed to move collaborative robots beyond static programmed tasks so manufacturers in dynamic, high-mix environments could deploy cobots for adaptive assembly and machine-tending work without complex manual reprogramming.
Reported impact
Universal Robots achieved up to 100x faster motion planning, enabled accurate 3D pose estimation with minimal training, shortened partner time to market for AI-powered cobot solutions, and opened new application categories and revenue streams across industrial automation.
Universal Robots built its AI Accelerator on the NVIDIA Isaac robotics platform to create a physical-AI stack for adaptive cobots. The platform runs on NVIDIA Jetson AGX Orin at the edge and combines NVIDIA Isaac Manipulator components such as cuMotion, FoundationPose, and FoundationStereo with Isaac ROS, Isaac Lab, and NITROS integration for ROS 2 environments. In production, the cobot stack handles real-time object recognition, 3D pose estimation, motion planning, and low-latency perception-to-control handoff without depending on the cloud. In development, Universal Robots uses Isaac Lab and simulated training with domain randomization to build reinforcement-learning policies that transfer to real assembly tasks without manual fine-tuning. This architecture lets the company package advanced AI perception and motion into an accessible cobot platform instead of forcing each customer to build custom robotics pipelines from scratch.
Sensors and cameras feed perception into the AI Accelerator on Jetson AGX Orin, Isaac Manipulator modules estimate pose and plan motion, ROS-based control loops execute adaptive robot actions, and simulation-trained policies from Isaac Lab move into real cobot tasks with minimal extra tuning.
"With the AI Accelerator, we're empowering our partners to bring AI-powered cobot solutions to market faster and with greater flexibility."
NVIDIA Jetson AGX Orin™ for low-latency, high-accuracy edge AI without cloud dependencies.
Universal Robots addressed these challenges by introducing the AI Accelerator, a platform for physical AI development that enables robots to perceive and interact with their environment and solve complex problems that were previously beyond their static capabilities.
Universal Robots integrated key components of the Isaac platform, such as AI models and libraries, into their AI Accelerator solution powered by NVIDIA Jetson AGX™.
Using NVIDIA Isaac Lab and Isaac ROS, Universal Robots has delivered a sim-to-real workflow for reinforcement learning training and real-world execution.
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
NVIDIA Sample Aug 25, 2026 Siemens Siemens needed to help industrial teams overcome fragmented engineering tools, labor shortages, and slow design and maintenance workflows so they could design, simulate, and operate complex products and factories faster… 10%–15% reductions in capital expenditure NVIDIA CUDA · NVIDIA DGX B200 · NVIDIA DGX H200 NVIDIA Sample Aug 24, 2026 Noble Machines Noble Machines needed to train, evaluate, and deploy a general-purpose industrial robot fast enough for real customer environments without relying on slow, physical-only testing or cloud-dependent inference. 15 people NVIDIA GeForce RTX 5090 GPUs · NVIDIA H100 GPUs · NVIDIA Isaac GR00T 1.7 NVIDIA Sample Aug 12, 2026 Pegatron Pegatron needed to improve factory planning, quality control, and operational efficiency across large manufacturing sites, but manual line design, fragmented agent development, scarce defect data, and labor-intensive… 10% decrease in operational effort NVIDIA AI Enterprise · NVIDIA Cosmos · NVIDIA DGX 