NXP releases robotics platform with open-source software and drone reference designs

NXP released a unified robotics platform on November 9, 2026, giving developers an open-source software development kit and drone reference designs available now. The company’s pitch is to shorten the work of connecting a robot’s computing, sensors and controls: it says pre-integrated software and validated subsystems can reduce integration and validation from months to days or weeks.

From separate parts to a shared development path

NXP had outlined the workflow in a November 5 product video: select prototype-ready hardware, download the validated software kit, then begin flight testing. That walkthrough presents the modular drone platform as the first application of the wider robotics platform, with reference designs for computing, flight control, motor control and networking.

The launch builds that approach around NXP’s Neural Axis Architecture. NXP describes it as a distributed foundation connecting sensing, AI processing, communications and real-time control—the functions that must respond on time as a machine operates. Rather than treating those functions as separate development projects, the platform supplies hardware and software intended to work together.

A developer could start by evaluating a motor-control subsystem already tested with the kit. NXP’s example then adds AI-based perception, flight control and navigation to build toward a complete autonomous drone. The intended progression is from one validated function to a larger system, rather than requiring a team to begin with the entire machine.

The software connects four drone building blocks

The modular software development kit, or SDK, includes supported drivers, software integration components, reference applications, configuration tools and reusable code. It also integrates with the open-source technologies PX4, Zephyr and ROS 2. Developers therefore get a common software package for the reference designs, with connections to established robotics software rather than an entirely separate environment.

The first hardware reference designs divide the drone’s work across four areas. Each is based on processors from NXP’s own portfolio:

• Autonomous computing and flight control: NavQ95, based on the i.MX 95 applications processor.
• Flight management: Tropic VMU/FMU, based on the i.MX RT1064 crossover microcontroller.
• Sensor networking: the 100Base-T1 Sensor Node, based on the MCX N947 microcontroller.
• Motor control: an AM32-compliant electronic speed controller, based on the MCX A13 microcontroller.

NXP says these designs and the SDK can deliver initial drone functionality within 30 minutes out of the box. That is a narrower claim than completing an autonomous drone in half an hour: the promised milestone is initial functionality. Its separate days-or-weeks estimate concerns the work of integrating and validating robotic subsystems.

You need efficient communication between computing elements, predictable behavior and extremely low latency.

Srdjan Kovačević, CEO and co-founder of Orqa, in NXP’s launch announcement

Drones now, other autonomous systems in 2027

Security and safety are also part of NXP’s design pitch. The reference designs include secure processing, predictable real-time control and safety-ready system architectures. Each includes the safety and security features of its primary processor, which can be enabled independently of the robotics-specific functions.

Jerome Legros, NXP’s vice president and general manager for robotics, linked those priorities to robots moving beyond controlled settings into places where people live and work. He identified security, functional safety, real-time control and reliability throughout a product’s life as foundations developers need to build trust.

The present release remains drone-focused. NXP says future reference designs will address service robots, autonomous mobile robots and other autonomous systems needing perception, navigation, motor control and connectivity. Additional support for other autonomous systems is expected in the second half of 2027. That is the next planned step, not a capability delivered across those robot categories at launch.