Lockheed Martin and Xaba cite 10x gain in robot drilling accuracy with xCognition
Lockheed Martin and Xaba reported that adding Xaba’s xCognition controller to an industrial robot improved drilling accuracy and consistency by 10x. The claim is tied to a two-phase evaluation that included drilling tests on an aluminum plate with specified positional tolerances. The result points to how robotics augmented with xCognition could take on operations typically kept on more expensive and less flexible CNC machines.
This story was produced through MarketScale. See how Industrial IoT teams put it to work with AI Visibility (GEO).
Key facts, context, and what it means.
Key takeaways
Lockheed Martin and Xaba reported a 10x accuracy-and-consistency improvement tied to a two-phase test that included drilling on an aluminum plate with specified positional tolerances.
If the goal is moving selected drilling or light machining off CNC, the first gating item is whether the controller enables path-mode accuracy under tolerance, not whether the robot can detect parts.
Moruzzi’s observation that most AI research emphasis is still on computer vision is a reminder to separate “perception AI” from “process AI” when writing automation requirements and budgets.
Free workspace
Turn your Industrial IoT expertise into content.
Record interviews, organize footage, and write with AI on a free trial of the MarketScale platform for qualifying companies. No demo required, no credit card.
That matters for any plant trying to shift selective drilling, light machining, or precision assembly away from fixed CNC machines and into flexible robotic work cells. The hard part is not getting a robot to move. It’s getting it to hold a path, under tolerance, repeatably.
What Lockheed and Xaba say improved, and how it was measured
Assembly Magazine and Robotics 24/7 both describe the collaboration as a two-phase evaluation in a typical aerospace robotic work cell. Phase 1 compared trajectory positioning performance with and without xCognition. Phase 2 ran drilling tests on an aluminum plate with specified positional tolerances.
Assembly Magazine and Robotics 24/7 reported that the xCognition-enhanced setup improved accuracy and consistency by a factor of 10. Assembly Magazine also said the result suggests robots augmented with the controller could perform operations that have been done on more expensive and less flexible CNC machines.
Aero is a proof point, but the real decision is CNC capacity planning
Why Xaba is calling this “physical AI,” not another vision stack
In a 2025 Machine Design Q&A, Xaba CEO Massimiliano Moruzzi argued that much of mainstream AI research is concentrated in computer vision and convolutional neural networks, while manufacturing outcomes often depend on physics, mechanics, and the equations that govern machine behavior. In that framing, xCognition is meant to be a physics-informed model that sits closer to the motion and process layer than to inspection cameras.
That framing aligns with how the World Economic Forum describes Physical AI. According to the World Economic Forum, training-based robotics can learn from simulated or real-world experience, handle more variation than rule-based systems, and virtualize training to reduce deployment time.
Spec questions to add to pilots and RFPs for precision robotic work
Sources
- Q&A: Integrating Physical AI with Physics and Mechanics of Industrial Machines ↗ · Machine Design
- What is physical AI -- and how is it changing manufacturing? ↗ · World Economic Forum
- Lockheed, Xaba Test Cognitive Autonomous Robots for Airframe Manufacturing ↗ · Assembly Magazine
- Xaba and Lockheed Martin Collaborate to Test Cognitive Autonomous Robots in Airframe Manufacturing ↗ · Robotics 24/7
Featured companies
Your experts belong here
Every story in MarketScale Industrial IoT starts with a company putting its controls engineers, plant-floor specialists, and integration partners on the record. Buyers are already reading this topic. The only question is whose experts they find.
Plant and controls buyers research deep before contact, and your engineers get to shape that research.
About the author
The MarketScale Newsroom reports on the companies, technologies, and trends shaping 16 B2B industries. It turns primary sources and expert commentary into clear, useful coverage for the people doing the work.
