NXP launches modular robotics platform for physical AI

NXP Semiconductors said on November 9, 2026, that its new modular robotics platform for physical AI can deliver initial drone functionality within 30 minutes out of the box.

The company said the package combines processors, open-source software and reference designs. NXP said pre-integrated software and validated subsystem interoperability can cut the months developers often spend integrating and validating robotic subsystems to days or weeks.

How does the platform shorten development?

NXP’s underlying Neural Axis Architecture is a distributed foundation for securely connecting sensing, real-time control, AI processing and communications across robotic systems.

NXP’s development path starts with a motor-control subsystem tested through the SDK. Developers can then add AI-based perception, flight control and navigation to build an autonomous drone. NXP said the workflow is intended to let teams scale from individual functions to complete physical AI systems within one validated development environment.

NXP’s immediate proposition is a common development route from motor control to autonomous flight.

What software comes in the Robotics SDK?

NXP says the Robotics SDK provides supported drivers, middleware integration, reference applications, configuration tools and reusable components.

• Supported development assets: Drivers, middleware integration, reference applications, configuration tools and reusable components.
• Robotics software: ROS 2 Jazzy, PX4 autopilot, CogniPilot Cerebri, Zephyr RTOS, NuttX and AM32 electronic-speed-controller firmware.
• Edge-AI software: eIQ Toolkit and Ara SDK.
• Middleware and connectivity: Zenoh and rmw_zenoh with zenoh-pico, MAVLink, uORB, DShot, 100BASE-T1 automotive Ethernet and RPMsg inter-core communication.
• Documented build environment: Linux with Docker, the west meta-tool, the Arm GNU toolchain, CMake and Ninja.
• Access condition: Developers need a Nexus user token to use prebuilt artifacts.

Taken together, the components support NXP’s goal of assembling robot functions inside one validated environment.

Which processors handle the drone’s core jobs?

NXP’s initial Modular Drone Platform divides four functions across four reference designs and processors.

• NavQ95: Autonomy compute and flight control, using the i.MX 95 applications processor.
• Tropic VMU/FMU: Flight management, using the i.MX RT1064 crossover microcontroller.
• 100Base-T1 Sensor Node: Intelligent sensor networking, using the MCX N947 microcontroller.
• AM32-compliant electronic speed controller: Motor control, using the MCX A13 microcontroller.

NXP said each design includes the safety and security enablement of its primary processor, which can be enabled independently of robotics-specific functions. The company described the reference designs as interoperable and validated across the complete drone system.

The four-design division separates autonomy, flight management, sensor networking and motor control across dedicated chips.

Why do security and latency matter?

Security and functional safety sit at the center of NXP’s design for robots operating around people. Jerome Legros, NXP’s robotics vice president and general manager, identified security, functional safety, real-time control and full-lifecycle reliability as the foundation for trusted physical AI systems.

Orqa CEO and co-founder Srdjan Kovačević framed the system-level requirements: “You need efficient communication between computing elements, predictable behavior and extremely low latency.”

NXP’s emphasis is trusted, predictable operation around people, not prototype speed alone.

Is the drone platform ready to ship?

NXP lists the platform, drone reference designs and SDK as available now. The company also labels the Modular Drone Platform and four associated boards as preproduction, and says the drone platform’s specifications and other information may change without notice.

The current package therefore gives developers access to software and reference resources while the underlying drone hardware remains nonfinal.

NXP’s status labels make the distinction clear: development resources are available now, but the hardware is not final.

What comes after drones?

NXP plans future reference designs for service robots, autonomous mobile robots and other autonomous systems. The company expects additional enablement in the second half of 2027. NXP said those machines share requirements for perception, navigation, motor control, connectivity and safe real-time operation.

The broader test will be whether hardware finalization and those additional designs extend the unified approach beyond the initial drone package.

For developers, the next measurable milestones are final hardware and second-half 2027 enablement beyond drones; until then, the package remains drone-first and partly preproduction.