Synaptics Introduces Astra SR200 MCU with Google Coral NPU
Synaptics has announced the Astra SR200 series xMCU, featuring an integrated Google Research Coral NPU, custom high-bandwidth DRAM, and up to 400 GOPS of AI performance. The chips combine Arm Cortex-M85 and Cortex-M52 CPU cores to enable multi-modal edge AI capabilities in microcontrollers. This release bridges the gap between traditional resource-constrained microcontrollers and heavy edge AI hardware, enabling advanced multi-modal AI workloads on low-power devices like wearables and smart sensors. By utilizing Google's open-source Coral NPU architecture, it helps standardize efficient edge AI design across the semiconductor ecosystem. The Astra SR200 series features scalable hundreds of megabytes of integrated custom DRAM and uses an Arm Cortex-M85 and Cortex-M52 CPU IP combination. Synaptics also introduced the Coralboard SR development board based on the highest-performing Astra SR250 variant to support diverse peripherals.
## BACKGROUND
Google's Coral NPU is an open-source, ultra-low-power machine learning accelerator based on the 32-bit RISC-V instruction set architecture, designed for efficient inference on wearable devices. Microcontrollers (MCUs) are compact integrated circuits designed to govern a specific operation in an embedded system, typically featuring limited memory and processing power compared to full microprocessors.