Renesas has introduced a hardware platform series for advanced driving assistance systems (ADAS) and automated driving systems. First product is the R-Car V3M System-on-Chip.
The System-on-Chip (SoC) launched at Devcon in Tokyo features two CAN FD interfaces and is dedicated for the use in smart camera applications, as well as surround view systems or even lidars. The product complies with the ISO 26262 functional safety standard and is equipped with a built-in image signal processor (ISP).
Vehicles in the autonomous driving era will be required to sense the environment, control the vehicle, and conduct synchronized communications with the cloud. A wide range of technologies is necessary to realize these functions and each technology needs to maintain high reliability to synchronize without any flaws. At the same time, these technologies are continuously advancing, which is why there is a growing demand for a total end-to-end solution. The Japanese company claims to be the only automotive semiconductor supplier to cover end-to-end solutions from secure cloud connectivity and sensing to vehicle control.
For the implementation of demanding algorithms, the introduced platform provides system manufacturers the option to select the most suitable IP cores, including hardware accelerators, offering functional safety. Functional safety is essential in achieving automated driving. Renesas has provided functional safety to the market since the launch of its first automotive safety MCUs in 2008. To address the requirements for high-level functional safety for ADAS and automated driving applications, Renesas supports several Asil (Automotive Safety Integrity Level) levels with its latest generation of SoC and MCU devices. In addition to Asil-C support with the R-Car V3M SoC, the chipmaker provides a range of Asil-B SoCs (R-Car M3 and R-Car H3), and Asil-D MCUs (RH850/P1X series).
Renesas has 195 suppliers worldwide in its R-Car Consortium and is continuously adding partners from the ADAS and automated driving areas to support the company’s hard platform. The development platform is open, allowing system manufacturers to choose where to enter, either building directly on Renesas innovations or working with ecosystem partners.
“ADAS and automated driving applications are driving the future of the automotive market. Renesas autonomy is a trusted, open and innovative platform that covers all needs from secure cloud connectivity, sensing, and cognitive over to vehicle control,” said Jean-Francois Chouteau, Vice President of Renesas Global ADAS Centre.
The R-Car V3M provides sufficient computing power to detect traffic signs, lanes, pedestrians, vehicles, and other obstacles in real time. Front cameras are mounted next to the windshield, where they are exposed to direct sunlight. As a result, low-power consumption of the SoC is crucial to minimize further heat dissipation in the smart camera housing. Securing functional safety is the final key element for front cameras to enable features such as automated emergency braking.
The SoC implements a top-notch performance computer vision platform using different accelerators, including a versatile pipeline engine and a computer vision engine, allowing the R-Car V3M to manage algorithms like optical flow, object detection and classification and convolutional neural networks. It integrates a dual Cortex-A53 for application programming and dual Cortex-R7 lockstep cores to run Autosar. The product achieves the hardware architectural metrics recommended for systems that require Asil-B and higher functional safety levels to act as a real-time core “safety island.”
The integrated ISP enhances the raw image quality from the camera sensor and makes the image ready for computer vision. The integration eliminates the need for an external ISP component in the front camera or in the sensor itself. The system implements a single DDR3L memory channel to further reduce cost. Other features include two CAN FD channels, video inputs and outputs, one Ethernet AVB interface as well as two Flexray ports. Samples will be available from December 2017. Mass production is scheduled to begin in June 2019.
News and reports