Nvidia Announces DRIVE Atlan, its Most Powerful Autonomous Vehicle Processor With 1,000 TOPS of Compute Power
【Summary】During today’s opening keynote of the annual GTC Technology Conference, NVIDIA founder and CEO Jensen Huang unveiled NVIDIA DRIVE Atlan, the upcoming generation of its AI-powered compute platform for autonomous vehicles. Atlan will achieve an unprecedented 1,000 trillion operations per second (1,000 TOPS) of performance to handle the large number of applications and deep neural networks that run simultaneously in autonomous vehicles.
Nvidia Corp kicked off its annual GTC Technology conference on Monday with some big new product announcements, including the latest version of its DRIVE autonomous vehicle platform that will transform self-driving vehicles into "supercomputers on wheels."
During today's opening keynote of the annual GPU Technology Conference, NVIDIA founder and CEO Jensen Huang unveiled NVIDIA DRIVE Atlan, the upcoming generation of its AI-powered compute platform for autonomous vehicles. Atlan centralizes the vehicle's entire computing infrastructure into a single system-on-a-chip (SoC).
Atlan will be able to handle the large number of applications and deep neural networks that run simultaneously in autonomous vehicles.
Nvidia said its next-generation Atlan platform will achieve an unprecedented 1,000 trillion operations per second (1,000 TOPS) of performance and an estimated SPECint benchmark score of more than 100, which is greater than the total compute power in most robotaxi vehicles in development today and four times faster than Nvidia's Orin processor.
Atlan will leverage NVIDIA's latest GPU architecture, new Arm CPU cores and deep learning and computer vision accelerators. The platform architecture provides ample compute horsepower for the redundant and diverse deep neural networks that will power future vehicles, while leaving headroom for developers to continuously add new features and improvements.
"To achieve higher levels of autonomy in more conditions, the number of sensors and their resolutions will continue to increase," Huang said. "AI models will get more sophisticated. There will be more redundancy and safety functionality. We're going to need all of the computing we can get."
Over the past several years, Nvidia has become a leading developer of hardware for autonomous vehicles around the world, due to the robust processing capabilities of its SoCs.
The company's DRIVE platform is a software-defined, end-to-end platform for the transportation industry that enables continuous improvement and continuous deployment through over the air updates. It delivers everything needed for the development of autonomous vehicles at scale.
Autonomous vehicles equipped with dozens of sensors and AI-powered software for navigation require an incredible amount of processing power and Nvidia's DRIVE family of products is designed to meet this need.
Automakers and startups around the world have turned to Nvidia to supply the hardware necessary to support the development of robust and safe autonomous driving technology powered by AI. Vehicles from Mercedes Benz, Toyota, Volvo, Audi and others are using Nvidia processors.
The open DRIVE Software stack allows developers build perception, mapping, planning, and driver monitoring capabilities. Using the power of GPUs and AI, developers can train deep neural networks that support these highly advanced systems.
Like the Nvidia DRIVE Orin, the next-gen platform is software compatible with previous DRIVE compute platforms, allowing customers to leverage their existing software investments across multiple product generations.
Atlan will come with Nvidia's BlueField data processing unit (DPU) for enhanced security
Atlan is also the first SoC to be equipped with Nvidia's BlueField data processing unit (DPU) for additional security, advanced networking and storage services. Nvidia's BlueField data processing unit will connect an autonomous vehicle directly to Nvidia's powerful data centers.
BlueField is a data center services accelerator on a chip, delivering up to 400 gigabits per second (Gb/s) of Ethernet and InfiniBand (IB) connectivity for both traditional applications and modern GPU-accelerated workloads while freeing host CPU cores to run applications instead of infrastructure tasks.
The NVIDIA BlueField DPU is designed to handle the complex compute and AI workloads required for autonomous vehicles. It's fully programmability to deliver "zero-trust" security to prevent data breaches and cyberattacks.
While vehicles are being packed with more computing technology, they're lacking the physical security that comes with data center-level processing, according to Nvidia. The company claims that Atlan is a technical marvel for safe and secure AI computing, fusing all of NVIDIA's technologies in AI, automotive, robotics, safety and BlueField data centers.
Data center development is critical to deploying robotaxis at scale. The ability to operate in thousands of conditions around the world requires intensive deep neural network training requiring massive amounts of data.
By combining data center and in-vehicle solutions, autonomous vehicle developers can create a continuous, end-to-end development cycle. As deep neural networks learn new capabilities in the data center, the validated algorithms can be pushed to the car's compute platform via an over the air software update.
Cars and trucks of the future will require an optimized AI architecture not only for autonomous driving, but also for intelligent vehicle features like speech recognition and driver monitoring systems. Upcoming software-defined vehicles will be able to converse with occupants answering questions, providing directions and even warn of road conditions ahead.
Nvidia is working with Mercedes Benz on the next generation of software-based vehicles that support over the air updates for the life of the vehicle. The advanced software-based vehicle architecture will be introduced beginning with 2024 model year vehicles, eventually rolling out to the entire Mercedes Benz fleet globally.
Nvidia and Mercedes Benz plan on developing the most sophisticated and advanced computing architecture ever deployed in the auto industry.
The new software-defined architecture will be built on Nvidia's Orin DRIVE platform. Eventually, every future Mercedes Benz model globally will come with the advanced and upgradable software-based architecture, from the entry-level A-Class to S-Class models.
"Every future Mercedes- Benz with the NVIDIA DRIVE system will come with a team of expert AI and software engineers continuously developing, refining and enhancing the car over its lifetime," said Huang last year when the partnership with Mercedes Benz was announced.
Nvidia hardware is also powering the autonomous driving systems for the new P7 sedan from Chinese electric vehicle startup and Tesla challenger XPeng.
Nvidia will first deliver its DRIVE Orin processors with vehicle production timelines starting in 2022. Orin will be followed by Atlan with sampling in 2023. It will be ready for production vehicles in 2025.
Originally hailing from New Jersey, Eric is a automotive & technology reporter covering the high-tech industry here in Silicon Valley. He has over 15 years of automotive experience and a bachelors degree in computer science. These skills, combined with technical writing and news reporting, allows him to fully understand and identify new and innovative technologies in the auto industry and beyond. He has worked at Uber on self-driving cars and as a technical writer, helping people to understand and work with technology.
Ford is Testing a New Robotic Charging Station to Assist Drivers of EVs With Disabilities
Ford Raises the Prices of the F-150 Lightning Electric Pickup Due to Rising Raw Material Costs
The BMW 7-Series to Feature HD Live Maps From HERE Technologies for Hands-Free Highway Driving in North America at Speeds up to 80 MPH
AutoX to Use the 'Eyeonic Vision Sensor' from California-based SiLC Technologies for its Robotaxi Fleet in China
LG Develops ‘Invisible’ Speaker Sound Technology That Could Revolutionize In-Vehicle Audio
Researchers at South Korea’s Chung-Ang University Develop a ‘Meta-Reinforcement’ Machine Learning Algorithm for Traffic Lights to Improve Vehicle Throughput
Zeekr’s New 009 Electric Passenger Van is the World’s First EV to Feature CATL’s Advanced ‘Qilin’ Battery With a Range of 510 Miles
Redwood Materials is Building an Electric Vehicle Battery Recycling Facility in South Carolina
- China’s Tesla Rival XPeng Unveils its New ‘S4’ 480kW Electric Vehicle Supercharger That Can Add 125 Miles of Range in Just 5 Minutes
- Ford CFO Claims Inflation Has Erased Mustang Mach-E Profits
- Amazon Customers in Texas and California Will Begin Receiving Packages via Prime Air Drones Later This Year
- Hyundai Motor Group Signs MoU with SK On Co. to Secure Batteries for EVs Built in the U.S.
- Tesla Rival XPeng and Alibaba Cloud Set Up China’s Largest Cloud-Based Computing Center to Train Machine Learning Models for Autonomous Driving
- Reservations for the Electric Fisker Ocean SUV Reach 50,000, The Company Shares New Details About its Second EV Being Built by Foxconn
- LG Develops ‘Invisible’ Speaker Sound Technology That Could Revolutionize In-Vehicle Audio
- Ford Motor Co Selects its Plant in Spain to Build 'Breakthrough' EVs for the European Market on a Next-Gen Architecture
- EV Startup NIO to Launch a New Mass-Market Brand to Rival Tesla with an Annual Capacity of 500,000 Vehicles a Year
- Ford Issues ‘Stop Sale’ of the Electric Mustang Mach-E Over Possible Loss of Propulsion While Driving