Nvidia, one of the leaders in the development of computing systems for self-driving vehicles, has revealed the architecture and underlying technology of ‘Parker’, its newest mobile SoC (System on a Chip) processor that will power the next generation of autonomous vehicles.
Earlier this year, Nvidia introduced its flagship Drive PX 2 platform, which uses two Parker processors and two Pascal architecture-based CPUs (Central Processing Units) to power deep learning applications. Since the launch, more than 80 auto makers, Tier 1 suppliers and university research centers around the world are now using the Drive PX 2 system to develop autonomous vehicles, including Volvo, which plans to road test Drive PX 2 systems in its XC90 SUVs next year. The Parker SoC delivers class-leading performance and energy efficiency, while supporting features important to the automotive market, such as deep learning, hardware-level virtualization for tighter design integration, a hardware-based safety engine for reliable fault detection and error processing, and feature-rich IO ports for automotive integration.
Built around Nvidia’s highest performing and most power-efficient Pascal GPU (graphics processing unit) architecture and the next generation of the company’s revolutionary Denver CPU architecture, Parker delivers up to 1.5 teraflops of performance for deep learning-based self-driving AI cockpit systems. Parker delivers 50-100% higher multi-core CPU performance than other mobile processors, due to its CPU architecture consisting of two next-generation 64-bit Denver CPU cores (Denver 2.0) paired with four 64-bit ARM Cortex A57 CPUs. A new 256-core Pascal GPU in Parker delivers the performance needed to run advanced deep learning inference algorithms for self-driving capabilities, and it offers the raw graphics performance and features to power multiple high-resolution displays, such as cockpit instrument displays and in-vehicle infotainment panels.
Working in concert with Pascal-based supercomputers in the cloud, Parker-based self-driving cars can be continually updated with newer algorithms and information to improve self-driving accuracy and safety. Parker includes hardware-enabled virtualization that supports up to eight virtual machines. Virtualization enables car makers to use a single Parker-based Drive PX 2 system to concurrently host multiple systems, such as in-vehicle infotainment systems, digital instrument clusters and driver assistance systems. Parker is a scalable architecture, so auto makers can use a single unit for highly efficient systems, or they can integrate it into more complex designs, such as Drive PX 2.
Drive PX 2 delivers an unprecedented 24 trillion deep learning operations per second to run the most complex deep learning-based inference algorithms, which deliver the supercomputer level of performance that self-driving cars need to safely navigate through all kinds of environments. To address the needs of the automotive market, Parker includes features such as a dual-CAN (controller area network) interface to connect to the numerous electronic control units in vehicles, and Gigabit Ethernet to transport audio and video streams. Parker is designed to support both decode and encode of video streams up to 4K resolution at 60 frames per second, enabling auto makers to use higher resolution in-vehicle cameras for accurate object detection, and 4K display panels to enhance in-vehicle entertainment experiences.