This page was machine-translated and may differ from the original. View original

Xilinx and Motovis Develop Automotive Front Camera Solution

Preferred source on GooglePublished2021.09.01 11:33
Front camera, a key element of LKA-AEB-ACC
Xilinx-Motovis, CNN-based camera system
Joint development. Supporting a market growing 20% annually.



Forward-facing camera systems are a key component of advanced driver assistance systems (ADAS), providing the advanced sensing capabilities required for critical safety features such as Lane-Keeping Assistance (LKA), Automatic Emergency Braking (AEB), and Adaptive Cruise Control (ACC).

Xilinx announced on the 1st that it has collaborated with Motovis, an embedded AI autonomous driving supplier, to develop a solution for vehicle recognition and control of automotive front camera systems. The two companies announced a solution that combines Xilinx's automotive-grade (XA) 'Zynq® SoC' platform and Motovis' CNN (Convolutional Neural Network) IP.
Xilinx and Motovis Develop Automotive Front Camera Solution
[Image = Xilinx]

This currently available solution leverages CNNs to support the parameters required for the European New Car Assessment Program (NCAP) 2022 requirements and enables fast and cost-effective image processing.

“The front-facing camera market is expected to grow at a rate of nearly 20 percent annually from 2020 to 2025,” said Ian Riches, vice president of global automotive at Strategy Analytics. “This solution meets the needs of automotive OEMs who need to address new standards and continually increasing requirements.”

This solution is scalable across the 16nm and 28nm XA Zynq SoC families, based on Motovis CNN IP. Optimized hardware and software partitioning capabilities, combined with a customizable CNN-specific engine hosting Motovis' deep learning network, enable cost-effective product implementations at various performance levels.

The images supported by the two companies' front camera solutions have a resolution of up to 8 megapixels. OEMs and Tier 1 suppliers can differentiate their designs and implement future-proof designs by layering their own functional algorithms on top of the Motovis perception stack.

Xilinx and Motovis will present details of their solutions at the Xilinx Adapt 2021 event, which will be held online on September 15th.
To request a correction, reply or follow-up report on this article, see how to file a request. Previously published statements are collected in corrections & replies.
이수민 Reporter

Comments