Renesas Electronics accelerates the development of ADAS systems with their new Perception Quick Start Software based on its R-Car V3H SoC. The solution provides reference software for COD, LiDAR obstacle detection and RFD – three essential recognition areas for sensor-based Level 2+ autonomous vehicle systems.
The SoCs provide a mixture of high computer vision performance and AI processing at low power levels, giving an optimised embedded solution for automotive front cameras in Level 2+ autonomous vehicles. To deliver state-of-the-art recognition technology, the company designed the SoCs with dedicated hardware accelerators for key algorithms comprising convolutional neural networks, stereo disparity, dense optical flow, and object classification. The new perception software offers an end-to-end pipeline reference for developers working with these complex yet cost-effective and power-efficient accelerators, which enables customers to kickstart their application design whether they are experts at using the accelerators or have restricted experience. The reference software covers input from a sensor or recorded data, all stages of processing and display output on a screen.
“Specialized hardware accelerators play an essential role in achieving the computer vision performance and accuracy required in embedded ADAS and autonomy applications while still meeting stringent in-vehicle power consumption limits; however, the complexity of these accelerators can present a steep learning curve,” said Tim Grai, Director of Automotive Advanced Systems Innovation Department, Renesas. “With the new Perception Quick Start Software, we are able to offer a set of application software along with the underlying primitives to simplify the use of these complex accelerators needed to achieve embedded ADAS.”
TU-Automotive Detroit 2019, June 5-6, Booth C190