Renesas Electronics Corporation has introduced a new device aimed at the high-volume visual artificial intelligence market

     Renesas Electronics Corporation has introduced a new device aimed at the high-volume visual artificial intelligence market, extending its RZ/V family of microprocessors. Similar to the high-end RZ/V2H, the new RZ/V2N MPU is equipped with the company's proprietary AI accelerator DRP-AI3, which boasts 10 TOPS/W power efficiency and up to 15 TOPS AI inference performance through its advanced pruning technology. With the latest addition of RZ/V2N, the RZ/V series is now extended to cover the full range of markets, from the low-end RZ/V2L (0.5 TOPS) to the high-end RZ/V2H (up to 80 TOPS).

     The new MPU is particularly smaller than the RZ/V2H, with a package area of just 15 square millimeters and a 38% reduction in installation area. The RZ/V2N inherits the advanced features of the family, combining high AI performance with low power consumption. These optimized features inhibit heat generation, eliminate the need for additional cooling fans, and reduce the size and cost of the embedded system. Developers can easily implement visual AI in a variety of applications, from AI cameras for traffic and congestion analysis in commercial facilities, to industrial cameras for visual inspection on production lines, and driver monitoring systems for behavioral analysis.

     Similar to the RZ/V2H, the new RZ/V2N is equipped with four best-in-class Arm Cortex-A55 CPU cores and one Cortex-M33 core, as well as a high-quality image signal processor (ISP), the Arm Mali-C55. RZ/V2N also has two MIPI camera interface channels, allowing two cameras to be connected to capture dual Angle images. Compared with the single camera system, the dual camera system greatly improves the spatial recognition performance, and realizes more accurate human motion line analysis and fall detection. In addition, a dual camera system captures images from different locations, allowing a single chip to count the number of vehicles in a parking lot and efficiently recognize license plates.

     "Since launching RZ/V2H last year to target next-generation robots that require visual AI and real-time control, we have had great interest in using the Renesas DRP-AI accelerator," said Daryl Khoo, vice president of embedded processing at Renesas Electronics. "With the addition of RZ/V2N, which shares the same heritage as RZ/V2H, we are expanding our scope to address mid-range applications, especially endpoint vision AI, that do not need to be implemented through power-hungry design." I am delighted that our customers will be able to choose the best AI MPU from Renesas that meets their system and budget requirements."

     "Visual AI applications in markets such as smart cities, enterprise and industrial are broad and diverse, but all require sustained performance and efficient processing," said Paul Williamson, senior vice president and general manager of Arm's iot business. Renesas' new RZ/V2N MPU leverages the leading capabilities of the Arm computing platform to meet the performance and efficiency needs of the next generation of visual AI use cases."

     RZ/V2N will be available along with the company's evaluation board suite and software development environment, as well as a variety of AI applications covering AI applications and more than 50 use cases available in the Artificial Intelligence SDK (GitHub). The SDK helps users without in-depth AI knowledge to quickly evaluate and develop AI applications, reducing time to market. In addition, the partner companies will provide SOM boards, SBC, camera modules and other products integrated with Renesas RZ/V2N. This reduces hardware design work, allowing developers to focus on AI application development and get their products to market quickly.