Many modern driver assistance systems rely on non-upgradable automotive electronic control units (ECUs) for data processing. For certain features, like automatic emergency breaking, that low-level processing is enough – but for autonomous features at level two and beyond, purpose-built processing systems are required to manage increased sensor data.
Read more: SAE International Releases Updated Visual Chart for Its “Levels of Driving Automation” Standard for Self-Driving Vehicles
DRIVE AutoPilot is a “Level 2+” automated driving solution that leverages the Nvidia DRIVE AGX Xavier system-on-a-chip (SoC) for wide array of advanced autonomous functions and safety features. Nvidia DRIVE AGX Xavier is capable of computing 30 trillion operations per second while consuming only 30 watts of energy – about half the energy of an incandescent consumer-grade lightbulb.
(Image source: Nvidia)
“This power and efficiency makes possible a much wider range of active safety features. It allows your car to run deep neural networks in parallel for surround perception, identifying and reacting to a variety of hazards all around the vehicle,” according to Danny Shapiro, Nvidia’s senior director of automotive solutions.
Nvidia DRIVE AGX Xavier runs a continuously updated Nvidia DRIVE Software stack, which automakers can use for object detection, traffic light, and sign recognition. It includes DRIVE AV autonomous driving software and DRIVE IX, which enables both driver monitoring and in-vehicle visualization to ensure the driver’s attention is on the road and act if the driver is drowsy or distracted.
(Image source: SAE International)
DRIVE AutoPilot also can drive previously driven routes through a feature called “My Route” when no HD map is available, enabling point-to-point automated driving.
“Powering AI-infused automated driving with the DRIVE AutoPilot also makes it easier to advance to higher levels of autonomy. DRIVE AGX Xavier can be upgraded to DRIVE AGX Pegasus — capable of 320 trillion operations per second — with few architectural changes,” says Shapiro.
Nvidia DRIVE AGX Pegasus delivers the performance of a trunk full of personal computers in an auto-grade form factor the size of a license plate. It currently provides Level 5 capabilities for robotaxis and DRIVE ecosystem partners are currently using DRIVE to bring vehicles with advanced AI safety features to the consumer market sooner.
Read more: Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles
ZF Friedrichshafen, a Germany-based developer of driveline, chassis, and automotive safety technology, unveiled its own scalable ProAI autonomous driving solution – built using Nvidia DRIVE Xavier processors Drive software and scheduled for production this year – earlier this week at CES 2019. The platform is designed for truck platooning and ranges in autonomous capability, from Level 2+ to Level 5.
Continental AG of Hanover, Germany also announced a production-level automated driving architecture that will bridge from “Premium Assist” features to future automated functionalities. The Level 2+ system uses Continental’s portfolio of radar, lidar, camera and Automated Driving Control Unit (ADCU) technology, powered by Nvidia DRIVE, and will go into production in 2020.
And in October, Volvo Cars of Gothenburg, Sweden announced the development of a Level 2+ systems on DRIVE AGX Xavier for vehicles slated for production in the early 2020s.
“Partners like these are using DRIVE AutoPilot to make cars safer now, and autonomous tomorrow. That makes building AI into vehicles a journey well worth starting today,” says Shapiro.
Bookmark http://www.sae.org/news to keep pace with the latest aerospace technology news and information.
Subscribe to SAE MOBILUS for access to more than 200,000 resources, including aerospace standards, technical papers, eBooks, magazines, and video.
William Kucinski is content editor at SAE International in Warrendale, Pa. Previously, he worked as a writer at the NASA Safety Center in Cleveland, Ohio and was responsible for writing the agency’s System Failure Case Studies. His interests include literally anything that has to do with space, past and present military aircraft, and propulsion technology.
Contact him regarding any article or collaboration ideas by e-mail at firstname.lastname@example.org.
Continue reading »