Improved deep learning performance SoC for latest NCAP requirements

Renesas Electronics Corporation has updated its state-of-the-art R-Car V3H SoC to achieve significantly improved deep learning performance for smart camera applications, comprising driver and occupant monitoring systems (DMS/OMS), automotive front cameras, surround view, and auto parking for high-volume vehicles up to Level 2+. The updated SoC connects sensor fusion on the real-time domain with up to ASIL C metrics and architecture optimised for smart computer vision. It provides OEMs and Tier 1s a high-performance, low-power solution that supports the newest NCAP 2020 demands and the roadmap to NCAP 2025 3 Stars at competitive system costs.

The updated SoC delivers four times the performance for CNN processing when compared to the previous version. It delivers up to overall 7.2 TOPS processing, incorporating all computer vision IPs while preserving low power consumption levels.

“The rising popularity of intelligent systems for applications, such as driver monitoring, park assist, driver recognition, and occupant sensing, is stoking higher demand for deep learning functions as automakers look for new ways to enable smart camera applications,” said Naoki Yoshida, vice president of Automotive Digital Products Marketing Division at Renesas. “The updated SoC provides R-Car V3H customers a flexible, easy, and cost-effective path to higher computer vision performance for next-generation front camera systems, and we are excited to continue bringing cutting-edge innovation to our customers with our growing platform of open and scalable SoCs for ADAS and highly automated driving.”