CPUs, FPGAs and software work to bolster automated driving

Author : Dan McNamara

05 January 2017

Intel’s FPGAs have been chosen by automotive supplier Denso to develop Advanced Driver Assistance Systems (ADAS) in automated vehicles.

The Consumer Electronics Show (CES) is abuzz every January with the newest technologies aimed at exciting and engaging the consumer. Of course, this includes the latest cars and automotive technology, and we are seeing strong demand for our field programmable gate arrays (FPGAs) in the automotive market. In fact, our FPGAs are solving intricate processing challenges in our customers’ automotive advanced driver assistance systems (ADAS) being introduced at CES. Some of the details include:

• Intel is unveiling Intel® GO™ for automated driving products, including the Intel GO In-Vehicle Development Platform for Automated Driving with Intel FPGAs sitting alongside either an Intel® Atom™ or Intel® Xeon® processor. These FPGA and CPU-based boards can serve automakers as they look for a rapid, yet reliable way to design ADAS systems, as outlined here by Doug Davis, vice president and general manager of Intel’s Automated Solutions Division (ASD).

• In addition, The Intel GO Automotive 5G Platform is based upon Arria 10 FPGAs and advanced RFIC, and gives developers the ?exibility to test di?erent prestandard 5G specifications.

• We’re also announcing the Arria 10 FPGA automotive grade solution, the highest performance processing automotive FPGA, which offers a full one speed grade performance advantage over competing devices, and auto manufacturers can run with this line of FPGAs for a head start on their ADAS systems.

• Additionally, we are sharing along with automotive supplier Denso a new stereo vision solution for automakers using the Intel Cyclone V SoC – enabling a new system that helps cars make braking decisions.

Specifically, the new automotive-grade Arria® 10GX mid-range FPGA for autonomous driving is up to 40 percent lower power than previous generation FPGAs and features the industry’s only hard floating-point digital signal processing (DSP) blocks. This enables developers to design to the strictest requirements. In addition, the Arria 10 automotive-grade FPGA offers more than 320 Gbps of DDR4 bandwidth and more than 200 Gbps of transceiver bandwidth. It also supports 12G serial protocols, 1.5K variable-precision DSP blocks and up to 24 transceivers.

Now, let me share a bit about our work with Denso, a forward-thinking and innovative automotive supplier with a global reach. Denso has been working for several years on enabling technology that makes cars “see” by using sensors that pull in visual data and process that data in rapid fashion. This kind of technology enables a car to help make decisions for the driver (autonomous driving) or with the driver (automated).

The technology behind Denso’s offering, which it calls the world’s smallest size form factor stereo camera system, is the Automated Emergency Braking ECU (AEB) for image recognition and processing that assists in automated emergency braking applicable for both pedestrians (up to 50 km/h) and obstacles (up to 80 km/h), brake assist systems and adaptive headlamp systems – all running on Intel’s Cyclone V SoC FPGA.

Officials with Denso’s Driving Assist and Safety Engineering department revealed to me that their technology teams worked to deliver an in-car stereo vision system to get the most accurate and reliable result for this automotive braking system. A lot of visual image processing is required, and they shared that Intel‘s FPGA proved to be the only suitable silicon solution available to perform such intensive computational processing while also meeting their power and cost requirements.

The stereo vision system is controlled by the Cyclone V SoC FPGA, which acts as a single-chip programmable high-performance computer to handle the intensive signal processing. Denso says this programmability is key because it lets automakers offer a differentiated solution from their competitors, as well as providing flexibility to vary the board across their own makes and models of cars.

In summary, we are seeing tremendous interest and adoption of Intel FPGAs across the automotive market including discrete ECU solutions, to the fully autonomous and centralized computing like the Intel GO platform. These are just a few examples of how Intel FPGA technology is helping advance driving systems as the industry moves toward the brave new world of ADAS.

Contact Details and Archive...

Print this page | E-mail this page