Renesas Targets High Volume Vision AI with DRP AI Core

Renesas Electronics has introduced the RZ/V2N, an integrated quad-core ARM microprocessor featuring its proprietary DRP embedded AI technology tailored for high-volume vision applications. The RZ/V2N is equipped with the Dynamically Reconfigurable Processor DRP-AI3, boasting 10 TOPS/W and AI inference performance of up to 15 TOPS thanks to its advanced pruning technology.

Compared to the high-end 80TOPS RZ/V2H, the RZ/V2N is notably smaller, with a package area measuring just 15 mm square, reducing the mounting area required by 38%. Its enhanced energy efficiency eliminates the need for additional cooling fans, thereby reducing the size and cost of embedded systems for vision AI across various applications.

The Edge AI core showcased in the RZ/V2N eliminates the need for a heatsink in robotics, showcasing its efficiency and compact design. Renesas is currently demonstrating the capabilities of the RZ/V2N at embedded world 2025 in Nuremberg, Germany, this week.

Featuring four ARM Cortex-A55 CPU cores and a single Cortex-M33 core, the RZ/V2N is complemented by a high-quality image signal processor (ISP) and Arm Mali-C55. Additionally, the RZ/V2N offers two channels of MIPI camera interfaces, enabling the connection of two cameras for capturing double-angle images.

A dual-camera system significantly enhances spatial recognition performance compared to a single-camera setup, facilitating more precise human motion line analysis and fall detection. This setup can capture images from different angles, allowing for efficient tasks such as counting cars in a parking lot and recognizing license plates using a single chip.

“Since the launch of the RZ/V2H last year targeting next-generation robotics requiring vision AI and real-time control, we have seen significant interest in utilizing the Renesas DRP-AI accelerator,” stated Daryl Khoo, VP of Embedded Processing at Renesas. “With the introduction of the RZ/V2N, which shares the same lineage as the RZ/V2H, we are expanding our reach to cater to mid-range applications, particularly endpoint vision AI that does not necessitate power-hungry designs.”

The RZ/V2N will be made available with the Renesas evaluation board kit and a software development environment, alongside various AI applications covering over 50 use cases provided in the AI Applications and AI SDK on GitHub. This SDK aids users with limited AI knowledge in swiftly evaluating and developing AI applications, thereby reducing time to market.

Furthermore, partner companies will offer a range of products incorporating the Renesas RZ/V2N, including SOM (system-on-module) boards, SBCs (single-board computers), camera modules, and more. This initiative aims to streamline hardware design work, enabling developers to concentrate on AI application development and expedite product launches.

Renesas has also developed the AI Camera reference design for applications in factories, public facilities, and commercial settings. This reference design comprises the RZ/V2N, power management IC, and real-time clock IC, serving as a blueprint for AI camera systems.

The RZ/V2N is set to be available from Renesas and its authorized distributors starting next week, offering a powerful and efficient solution for vision AI applications across diverse industries.