MIPS and INOVA launch physical AI reference platform for robotics
MIPS and Inova Semiconductors have unveiled a new reference platform aimed at accelerating the development of robotics and physical AI systems. The companies say the platform is designed to simplify the architecture of advanced robotic systems — particularly humanoids — while reducing size, power consumption, and development complexity.
For engineers and developers working on robotics, edge AI, and embedded systems, the platform could offer a practical blueprint for building scalable robot control architectures. With growing demand for intelligent machines at the edge, the collaboration targets faster development cycles and lower system cost for next-generation robotics platforms.
A reference architecture for robotics control
The collaboration combines technologies from MIPS, Inova Semiconductors, and manufacturing partner GlobalFoundries to create a robotics control reference platform optimized for physical AI workloads. At its core, the system implements a “sense-think-act-communicate” architecture designed to handle real-time control loops alongside secure AI processing.
The platform integrates Inova’s APXpress high-speed interface technology with MIPS’ RISC-V processor portfolio, including the Atlas M8500 high-performance microcontroller processor and the Atlas S8200 AI processor. These are complemented by MIPS mixed-signal technologies and implemented on GlobalFoundries’ FDX process platform, which is designed to support low-power operation.
Together, the companies say this approach enables mixed-criticality computing in robotics applications, allowing real-time motion control, AI inference, and secure communications to operate within a unified architecture.
“Together with INOVA, we’re delivering a Physical AI reference platform that simplifies robot design, reduces BOM cost, and gives builders an open, standards-based path to create whole product families with low latency and functionally safe connectivity,” said Sameer Wasson, MIPS CEO. “Robotics is moving rapidly and the leaders will scale quickly and cost-effectively. By pairing INOVA’s high-speed communication links with MIPS’ open RISC-V compute and mixed signal technologies, this scalable reference platform turns ‘sense-think-act-communicate’ into a Physical AI building block that lowers risk, lowers cost, and accelerates time to market.”
Scaling humanoids and advanced robotics
The reference design focuses on advanced robotics use cases such as multi-axis motion control in robotic arms and humanoid systems. By aggregating multiple data interfaces and supporting various network topologies, the platform aims to simplify the communication backbone required for complex robotics systems.
Inova’s expertise in automotive zonal architectures also plays a role in the design. Moreover, the company’s APXpress data backbone supports hundreds of independent data channels with minimal latency, which could help robotics developers handle large sensor and actuator networks more efficiently.
“Advanced humanoids demand secure, deterministic connectivity and a scalable control backbone. INOVA together with GF & MIPS, we’re giving robot makers a zonal, RISC-V-based blueprint that cuts complexity and cost to help scale humanoids and advanced robotics from prototype to production faster,” said Robert Isele, CEO at INOVA. “The creation of a reference zonal architecture for advanced robotics will enable simpler and faster creation of humanoid and other robotic form factors.”
To support early development, the companies are offering access to the platform through the MIPS Atlas Explorer environment, a simulation-based hardware/software co-design platform that allows developers to begin optimizing AI and control software before hardware is available.
MIPS and GlobalFoundries are demonstrating the platform at Embedded World 2026 in Nuremberg, Germany.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
