Stmicroelectronics Unveils First Automotive MCU With Built‑In AI Acceleration
STMicroelectronics has announced the Stellar P3E, the first automotive microcontroller (MCU) to feature integrated artificial intelligence acceleration, a development aimed at supporting the next generation of software‑defined vehicles.
The Stellar P3E is intended to simplify the consolidation of multiple vehicle functions into X‑in‑1 Electronic Control Units (ECUs), an approach that manufacturers are adopting to reduce cost, wiring, weight and system complexity.
“Stellar P3E sets a new benchmark for automotive electrification by combining high-performance real-time control and edge AI in a single device that meets the highest automotive safety levels,” said Luca Rodeschini, Group Vice President and General Purpose and Automotive Microcontrollers Division General Manager at STMicroelectronics. He added that the device’s processing performance, memory capacity and sensing features support emerging applications, including virtual sensors, and help carmakers develop “safer, more efficient and more responsive driving experiences.”
A central feature of the new MCU is the inclusion of ST’s Neural‑ART Accelerator — the first embedded neural network accelerator to appear in an automotive MCU. The component is built around a dedicated neural processing unit and an optimised data‑flow architecture that supports real‑time AI workloads.
STMicroelectronics says the P3E can execute inference tasks at microsecond speeds, delivering up to 30 times greater efficiency than conventional MCU cores. This performance enables low‑power, always‑on AI features suited to predictive maintenance, real‑time system monitoring and advanced sensing functions. Such capabilities could, for example, improve charging performance in electric vehicles or expedite deployment of new software features throughout a vehicle’s lifecycle.
By embedding AI acceleration directly at the MCU level, vehicle manufacturers can implement new capabilities without relying heavily on additional sensors or high‑power central processors.
“Shifting neural processing from centralised hubs to the edge of the vehicle enables sub millisecond-level decision-making, which is essential for the next generation of in-vehicle intelligence,” said Greg Basich, Associate Director at Counterpoint Research. He noted that integrated AI hardware allows manufacturers to offer predictive performance monitoring and virtual sensor applications “without the cost and thermal burden of a full-scale SoC.”
The Stellar P3E incorporates ST’s xMemory technology, a non‑volatile memory based on phase‑change memory (PCM)
. The company says this approach delivers twice the density of standard embedded flash and is qualified for automotive requirements. The extensible design is intended to support the growing software footprint of software‑defined vehicles, allowing new features and updates without hardware redesigns.
The new MCU is fully compatible with the ST Edge AI Suite, which provides workflows from dataset creation through to on‑device deployment. NanoEdge AI Studio, now extended across the Stellar family, is included within the suite. The P3E also integrates with Stellar Studio, the company’s development environment for automotive engineers, creating a combined hardware‑software ecosystem for building and deploying edge AI applications.
STMicroelectronics plans to begin production of the Stellar P3E in the fourth quarter of 2026.
#STMicroelectronics#pinterest #GenZGrowth #DigitalGrowth #InvestorUpdate
#printpublication #printmagazine #Modernautomotiveindia