San Jose, California&Shanghai - (Business Wire) - Xilinx, Inc. (NASDAQ: XLNX), a leader in adaptive comput
San Jose, California&Shanghai - (Business Wire) - Xilinx, Inc. (NASDAQ: XLNX), a leader in adaptive computing, announced today that two companies are collaborating to develop a solution that will integrate Xilinx Au with embedded artificial intelligence and autonomous driving
Time: November 18, 2023 19:18:55 Author: Click: 100
San Jose, California&Shanghai - (Business Wire) - Xilinx, Inc. (NASDAQ: XLNX), a leader in adaptive computing, announced today that two companies are collaborating to develop a solution for Xilinx Automotive (XA) Zynq, an embedded artificial intelligence autonomous driving provider ® The System on Chip (SoC) platform and Motovis' Convolutional Neural Network (CNN) IP are applied in the automotive market, specifically for vehicle perception and control of forward facing camera systems. This solution is based on Xilinx's enterprise plan, providing customers with a powerful platform to enhance and accelerate development.
The forward facing camera system is a key element of advanced driving assistance systems, as they provide advanced sensing capabilities required for safety critical functions, including lane keeping assistance (LKA), automatic emergency braking (AEB), and adaptive cruise control (ACC). The solution has now been launched, which utilizes convolutional neural networks to achieve a cost-effective combination of low latency image processing, flexibility, and scalability, supporting a range of parameters required by the European New Car Assessment Program (NCAP) 2022.
Ian Riches, Vice President of Global Automotive Business at Strategy Analytics, said, "This collaboration is an important milestone for the forward camera market as it will allow automotive OEMs to innovate faster. The future camera market has huge growth opportunities, and we expect sales to increase by nearly 20% year-on-year from 2020 to 2025. Xilinx and Motovis are partnering to launch highly optimized hardware and software solutions that will greatly meet the needs of automotive OEMs, especially with the emergence of new standards and the continuous growth of demand
The front camera solution can be extended to 28 nanometer and 16 nanometer XA Zynq SoC series using Motovis' CNN IP, which is a unique combination of optimized hardware and software partitioning capabilities and a customizable CNN dedicated engine hosting Motovis deep learning networks, providing cost-effective products at different performance levels and prices. This solution supports image resolutions of up to 8 million pixels. Original equipment manufacturers and first tier suppliers can now, for the first time, build their own functional algorithms on Motovis' perception stack to differentiate and adapt to future designs.
We are delighted to launch this new initiative with Xilinx and bring our CNN forward facing camera solution to the market. Customers who design systems that support AEB and LKA functionality need to perform efficient neural network processing within one SoC, allowing them to flexibly and easily implement future features, "said Dr. Huazheng Yu, CEO of Motovis. With Motovis' customizable deep learning network and the ability of Xilinx Zynq platform to host CNN specific engines, we are helping design for the future to meet customer needs“
Market forces continue to drive the adoption of forward facing camera systems to comply with global government directives and requirements from Consumer Watch organizations, including the European Commission's General Safety Regulation, the National Highway Traffic Safety Administration, and NCAP. These three companies have issued formal instructions or strong guidance on the implementation of LKA and AEB in new cars produced by car manufacturers from 2020 to 2025 and beyond.
By expanding our XA products with a comprehensive solution for the front-end camera market, we have put cost optimized high-performance solutions into the hands of our customers. Willard Tu, Senior Director of Xilinx Automotive, said, "We are pleased to achieve this and drive the industry forward. Motovis' expertise in embedded deep learning and how they optimize neural networks to address the enormous challenges of forward facing camera perception puts us in a unique position to win market share and accelerate time to market for our OEM customers
Xilinx and Motovis will participate in the Xilinx adapter virtual event on September 15, 2021. Adapt 2021 will host executive keynote speeches attended by partners and clients, as well as a series of over 100 presentations, forums, product training, and labs aimed at helping users unleash the value of adaptive computing.
