NVIDIA unveils new centralized computing technology for autonomous vehicles

NVIDIA unveils new centralized computing technology for autonomous vehicles

NVIDIA has unveiled DRIVE Thor, a centralized car computer designed to make the driving experience safer and easier for self-driving vehicles.

The new offering combines infotainment, automated driving and parking in a single, cost-effective system.

DRIVE Thor is claimed to achieve up to 2,000 teraflops of performance and unifies intelligent functions including automated and assisted driving, parking, driver and occupant monitoring, digital instrument cluster, in-vehicle infotainment (IVI) and rear entertainment.

It also includes the next-generation superchip and NVLink-C2C chip interconnect technology, which comes with advanced AI capabilities that were first introduced in the NVIDIA Hopper multi-instance GPU architecture.

The company also emphasizes efficiency and effectiveness. DRIVE Thor with MIG support for graphics and compute enables IVI and advanced driver assistance systems to perform domain isolation, allowing critical concurrent processes to run without interruption.

“Advances in accelerated computing and artificial intelligence are advancing at the speed of light,” said Jensen Huang, Founder and CEO of NVIDIA.

“DRIVE Thor is the superhero of centralized computing, with lightning-fast performance to deliver software-defined supercomputers on wheels, continuously scalable, safe and secure.”

As for the market, Geely-owned automaker ZEEKR has announced that it will integrate DRIVE Thor into its centralized vehicle computer for next-generation smart electric vehicles, starting production in early 2025.

“ZEEKR users demand a luxury experience that includes the latest technology and security features,” said An Conghui, CEO of ZEEKR.

“NVIDIA DRIVE Thor will support our mission to deliver cutting-edge technology that meets our customers’ needs and ensures that ZEEKR remains at the forefront of tomorrow’s innovations.”

Typically, with similar technology, dozens of electrical control units are distributed throughout a vehicle to power individual functions.

With DRIVE Thor, manufacturers can consolidate many functions onto a single system-on-chip (SoC), alleviating supply constraints and simplifying vehicle design development, resulting in significantly lower costs, less less weight and fewer cables.

Another advantage of DRIVE Thor is its 8-bit floating point (FP8) capability. Often, developers lose the accuracy of the neural network when changing from 32-bit FP data to 8-bit integer format. NVIDIA claims DRIVE Thor has 2,000 teraflops of FP8 precision, enabling transition to 8-bit without sacrificing precision.

DRIVE Thor is also the first AV platform to integrate an Inference Transformer Engine, a new component of Tensor Cores into NVIDIA GPUs.

With this engine, DRIVE Thor can accelerate the inference performance of Transformer deep neural networks up to 9 times, which is paramount for supporting complex AI workloads associated with self-driving.

The DRIVE Thor SoC and AGX board is developed to comply with ISO 26262 standards, and the software stack is designed for ISO 26262 and ASPICE compliance.

“The shift to software-defined vehicles with centralized electronic architectures is accelerating, driving a need for more powerful and energy-efficient computing platforms,” says Sam Abuelsamid, Principal Research Analyst, Guidehouse Insights.

“The virtualization, high-speed data transfer, and massive processing performance of NVIDIA DRIVE Thor can enable safer vehicles, better user experiences, and new potential revenue streams.”

The technology will be available for automakers’ 2025 models and will accelerate production roadmaps by bringing higher performance and advanced features to market in the same time frame.

#NVIDIA #unveils #centralized #computing #technology #autonomous #vehicles

Leave a Comment

Your email address will not be published.