Updated R-Car V3H aids deep learning performance for smart camera apps
Renesas Electronics has updated its R-Car V3H system-on-chip (SoC) to improve deep learning performance for smart camera applications, such as driver and occupant monitoring systems, automotive front cameras, surround view and auto parking for high-volume vehicles up to Level 2+.
The updated SoC performs sensor fusion using an architecture optimised for smart computer vision and supporting safety certification up to ASIL C for the real-time control domain. The device offers original equipment manufacturers (OEMs) and Tier 1s a high-performance, low-power solution that supports the latest NCAP 2020 requirements.
Building on the recognition technology introduced with the R-Car V3H in 2018 which integrated IP for convolutional neural networks (CNN), the updated R-Car V3H preserves full hardware and software compatibility but delivers four times the performance for CNN processing compared to the earlier version and achieves up to overall 7.2 TOPS throughput, including 3.7 TOPS for CNN functions, with optimised performance-to-power balance.
By supporting safety goals up to ASIL C in the real-time control domain, the design of the R-Car V3H removes the need for an external safety microcontroller (MCU) to manage sensor fusion and final-decision actions. Designed to support ASIL B goals in the sensor layer and on-chip applications processors, the SoC can be incorporated into systems that require ASIL D certification at the system level.
The R-Car V3H features a suite of IPs that support the perception stack, and enable sensor fusion involving radar, Lidar and cameras, as well as integrating a full set of automotive peripherals including CAN, Ethernet AVB, and Flexray.
Mass production of the updated R-Car V3H is planned for first quarter of 2022.
A global leader in microcontrollers, analogue, power, and SoC products, Renesas provides solutions for a broad range of automotive, industrial, Infrastructure, and IoT applications.