Platform innovates by providing a cutting-edge development environment

Renesas Electronics Corporation has announced the R-Car V4H SoC for central processing in ADAS and AD solutions. The platform achieves deep learning performance of up to 34TOPS, allowing high-speed image recognition and processing of surrounding objects by automotive cameras, radar, and lidar.

The platform allows market-leading performance per watt by carefully combining best-in-class IP and expert HW optimisation. It targets the highest volume zones of automated driving: Level 2+ and Level 3.

Thanks to a high level of integration, it enables customers to develop cost-competitive, single-chip, ADAS ECUs. These control units can support driving systems appropriate for automated driving Levels 2+ and Level 3, including full NCAP 2025 features. The R-Car V4H supports surround view and automatic parking functions with impressive 3D visualisation effects such as realistic rendering.

“We have seen excellent customer response to our R-Car V3H and V3M with high volume production, and we are pleased to extend our offering with the R-Car V4H,” said Naoki Yoshida, vice president, Automotive Digital Products Marketing Division at Renesas. “As Renesas expands its scalable R-Car portfolio, we expect to see cutting-edge ADAS adopted in all vehicle types, from mid-range to entry-level.”

Regarding ISO 26262 functional safety, the SoC development process targets ASIL D systematic capability for all safety-relevant IP. The signal processing portion of the platform is expected to achieve ASIL B and D metrics for the real-time domain.

Furthermore, the company provides a dedicated power solution for it based around the RAA271041 pre-regulator and the RAA271005 PMIC. This allows a highly reliable power supply for the R-Car V4H and peripheral memories from the 12V supply of the vehicle battery. These features allow low power operation while aiming at ASIL D compliance for systematic and random hardware faults at extremely low BOM cost. This assists to minimise the effort of hardware and software development while lowering design complexity, price, and time to market.

An SDK is also offered to perform faster and easier initial device evaluation and software development, incorporating deep learning. The SDK provides full functionality for machine learning development and optimisation of embedded systems for power efficiency, performance, and functional safety. Complete simulation models are available, and the company's operating system agnostic software platform allows easier development of the software-defined car.

When developoing from scratch, Fixstars’ Genesis platform allows engineers to evaluate R-Car from anywhere through their cloud solution and deliver quick and easy CNN benchmark results.