Search

Automated Driving Qualcomm’s new platform for all levels of automated driving, Snapdragon Ride

| Author/ Editor: Luke James / Erika Granath

In early January, Qualcomm unveiled its new automotive platform called “Snapdragon Ride” – consisting of a Snapdragon Ride system-on-chip, accelerator, and autonomous slack – that is designed to power all levels of automated driving.

Qualcomm says that the Snapdragon Ride platform is designed to support the automated vehicle industry’s top three segments.
Qualcomm says that the Snapdragon Ride platform is designed to support the automated vehicle industry’s top three segments.
(Source: Qualcomm)

Although Qualcomm may be best known in the smartphone space, it also has a family of its own automotive products – including semiconductors for telematics systems. Snapdragon Ride is the company’s latest and perhaps most notable automotive solution which is set to make waves in the already crowded ADAS and automated driving industries. It will also see Qualcomm competing with the likes of Nvidia and Intel.

Scalable across all levels of autonomy

With its new platform, Qualcomm is taking aim at the autonomous vehicle industry and hopes to accelerate the deployment of vehicles with a centralized architecture. This will create a comprehensive all-in-one platform that enables seamless updates to a vehicle’s features and functionality.

Snapdragon Ride is designed to scale across all levels of vehicle autonomy, from Level 1 vehicles that offer basic driving assistance through to Level 5, completely autonomous self-driving systems. It combines hardware (comprised an advanced system-on-chip), software, open stacks, development kits, tools, and a partner ecosystem.

Bringing new compute capabilities to mainstream cars

Qualcomm says that the Snapdragon Ride platform is designed to support the automated vehicle industry’s top three segments. In addition to handling the hardware requirements for active safety ADAS such as lane keeping and automatic emergency breaking, it can also manage more intensive hardware requirements seen in Level 2+ vehicles, such as self-parking.

Qualcomm says that the Snapdragon Ride platform is designed to support the automated vehicle industry’s top three segments.
Qualcomm says that the Snapdragon Ride platform is designed to support the automated vehicle industry’s top three segments.
(Source: Qualcomm)

The platform is built on modular multi-core CPUs, AI, computer vision engines, and a GPU. It is also thermally efficient, solving the problem that sees most high-performance centralized systems using complex and expensive liquid cooling systems because they burn through lots of power. According to Qualcomm, Snapdragon Ride can deliver over 700 Tera Operations Per Second (TOPS) for Level 4/5 driving at 130 W. This enables simpler, cheaper, and more reliable designs that are passively air-cooled and should bring new compute capabilities to mainstream cars.

“We are pleased to be introducing our first-generation Snapdragon Ride platform, which is a highly scalable, open, fully customisable and highly power optimised autonomous driving solution designed to address a range of requirements from NCAP to L2+ Highway Autopilot to Robo Taxis. Combined with our Snapdragon Ride Autonomous Stack, or an automaker or tier-1’s own algorithms, our platform aims at accelerating the deployment of high-performance autonomous driving to mass market vehicles,” said Nakul Duggal, senior vice president, product management, Qualcomm Technologies.

“We’ve spent the last several years researching and developing this new autonomous platform and accompanying driving stack, identifying challenges and gathering insights from data analysis to address the complexities automakers want to solve.”

Snapdragon Ride is expected to be available for pre-development to automakers and Tier-1 suppliers in the first half of 2020. Snapdragon Ride-enabled vehicles are expected to be in production in 2023.

Curious to know how Snapdragon Ride works in practice? Have a look at Qualcomm's demo video:

This article is protected by copyright. You want to use it for your own purpose? Contact us at support.vogel.de (ID: 46397370)