Incredibly scalable in-vehicle computing gives developers more flexibility.
A software development kit maximizes hardware capabilities.
A 5G-ready platform speeds innovation for new use cases.
A robust data center powers artificial intelligence (AI) and handles unprecedented amounts of data.
Solutions for Autonomous Driving, from Car to Cloud
Autonomous driving will change lives and societies for the better, resulting in fewer accidents, greater mobility, and more efficient traffic flow. But it will also create massive amounts of data that need to be captured, analyzed, moved, processed, and managed.
Autonomous driving on a global scale requires unprecedented compute in the vehicle and data center, sophisticated connectivity, and support for a software-defined infrastructure that can flex and scale with increasing levels of automation.
With Intel® GO™ automotive solutions, Intel brings its deep expertise in compute, connectivity, and the cloud to the automotive industry.
Intel GO Automotive Development Platforms
As driving becomes more automated, the vehicle must be able to visualize the road ahead, evaluate countless possible scenarios, and choose the best sequence of actions. It must process millions of data points every second and quickly respond to a constantly changing environment. This requires a tremendous amount of both parallel and sequential computing.
Intel GO automotive development platforms offer a flexible and scalable architecture of CPUs, FPGAs, and other accelerators. This provides a unique and optimized blend of parallel and sequential processing—ideal for partitioning automated workloads into the most efficient compute type. With a combination of Intel Atom® and Intel® Xeon® processors for automotive and Intel® Arria® 10 FPGAs, Intel delivers a diverse range of computing elements that can accommodate designs that may change long after hardware and vehicle design decisions have been made.
Ideal Balance of Sequential and Parallel Computing
The compute required for autonomous driving can be divided into three intertwined stages: sense, fuse, and decide. Each stage requires different levels and types of compute performance. In the first stage, the vehicle collects data from dozens of sensors to “see” its surroundings. During the second stage, data is correlated and fused to create a model of the environment. Finally, the vehicle must decide how to proceed. Intel delivers a flexible architecture that gives system designers a blend of parallel and sequential processing to support all three stages, with an optimized combination of power efficiency and performance.
Incredibly Scalable Development Platforms
The Intel® GO™ development platform for autonomous driving, including both Intel Atom and Intel Xeon processor versions, makes it easier for developers to build, evaluate, benchmark, and optimize solutions, from advanced driver assistance systems (ADAS) to fully autonomous vehicles. These platforms jump-start development, enable flexibility in design, and speed time to market. Both platforms include Intel Arria 10 FPGAs to speed production and come with a set of sample applications, run times and libraries, and middleware. In addition, they provide building blocks to enable developers to deliver functional safety and security to platforms.
Intel GO Automotive Software Development Kit (SDK)
The software stack within autonomous driving systems must be able to efficiently handle demanding real-time processing requirements while minimizing power consumption. The Intel GO automotive SDK helps developers and system designers maximize hardware capabilities while speeding the pace of development with a variety of tools:
- Computer vision, deep learning, and OpenCL™ tool kits to rapidly develop the necessary middleware and algorithms for perception, fusion, and decision-making.
- Sensor data labeling tool for the creation of “ground truth” for deep learning training and environment modeling.
- Autonomous driving-targeted performance libraries, leading compilers, performance and power analyzers, and debuggers to enable full stack optimization and rapid development in a functional safety compliance workflow.
- Sample reference applications, such as lane change detection and object avoidance, to shorten the learning curve for developers.
Intel GO Automotive 5G Platform
To confidently support vehicle-to-everything (V2X) communications, over-the-air updates, and new in-vehicle experiences, providers will need increasingly higher data transfer speeds, as well as faster response times—not just in seconds, but in milliseconds. The Intel GO automotive 5G platform offers the industry’s first 5G-ready platform for the automotive segment. This platform allows automakers to develop and test a wide range of use cases and applications for 5G:
- High-definition (HD) map downloads in real time
- HD content for in-vehicle infotainment (IVI)
- Over-the-air firmware and software updates
- Sensor data uploads from the vehicle for machine learning
- Use cases leading to safety, smart intersections, and cooperative driving
Intel® Technologies for the Data Center
High-performance computing in the car is essential to making immediate driving decisions. However, the data center is responsible for critical artificial intelligence (AI) simulation and ongoing training. The data generated by autonomous vehicles will serve as a new kind of currency, opening the door for the automotive ecosystem to act on emerging business opportunities. The greatest opportunity lies in AI, as machine learning and deep learning will enable autonomous driving models. In addition, data about traffic, roads, and users can be used to create new applications and better experiences.
Intel provides extensive data center capabilities and expertise to support these demanding workloads. Intel technologies for the data center support Intel GO automotive solutions with full scalability to continuously store and manage unprecedented volumes of data and enable cloud services.
- Sophisticated hardware, based on Intel Xeon and Intel® Xeon Phi™ processors, delivers the high-performance computing needed to support AI and other intensive workloads.
- Platform services, including database services, distributed compute engines, and frameworks for machine learning and deep learning, offer specialized support for autonomous driving.
- Functional applications and capabilities for autonomous driving are optimized to run most efficiently on data center infrastructure.
Future Intel Automotive Solutions
As part of Intel’s vision to accelerate the adoption of autonomous driving, Intel is planning to introduce an extensive roadmap of new solutions, from processors optimized for automotive and autonomous driving, to acceleration capabilities for computer vision and deep learning. In addition, Intel will continue to provide new automotive reference platforms to enable carmakers and automotive suppliers to accelerate innovation and speed time to market.
1. “Data Is the New Oil in the Future of Automated Driving.” Intel, Nov. 2016, newsroom.intel.com/editorials/krzanich-the-future-of-automated-driving/.