Follow
Subscribe

Volvo is Building a 225 Million Gigabyte Data Center to Collect & Process Real-Time Vehicle Data to Improve Safety

Home > News > Content

【Summary】​Swedish automaker Volvo Cars held a Tech Day on Wednesday where company executives shared a roadmap of the company’s future plans, including a switch to all-electric and software-based vehicles by 2030. These plans include having ten of thousands of connected cars on the road in the next decade traveling millions of kilometers and continuously sharing data with the automaker.

Eric Walz    Aug 07, 2021 10:00 AM PT
Volvo is Building a 225 Million Gigabyte Data Center to Collect & Process Real-Time Vehicle Data to Improve Safety
The Volvo Recharge Concept.

Swedish automaker Volvo Cars held a Tech Day on Wednesday where company executives shared a roadmap of the company's future plans, including a switch to all-electric and software-based vehicles by 2030.

A big part of Volvo's future plans include having ten of thousands of software-powered connected cars on the road in the next decade traveling millions of kilometers and continuously sharing data with the automaker. 

All of these software-based vehicles will be like rolling smartphones that can be updated OTA.

The crowd-sourced data collected from Volvo vehicles will include continuous inputs from vehicle sensors that monitor the environment, including high-resolution lidar data used for autonomous driving. Allowing customers to share vehicle data will help Volvo continuously make software improvements to its cars, including the advanced safety systems, including autonomous driving systems.

For privacy concerns, Volvo customers will be able to choose whether to share their vehicle data or not. However, Volvo says that all data collected will be aggregated to protect customer privacy.

Volvo said its engineers are able to validate and verify autonomous driving capabilities much faster using real-world vehicle data, rather than on a closed track using data from a few test vehicles. Any software improvements made can then be pushed to the fleet of Volvo connected vehicles via an OTA update. 

Autonomous driving technology relies heavily on machine learning algorithms to make driving decisions. However, training these algorithms requires a massive amount of real world data, which is known as "training data." The fleet of connected Volvo vehicles would continuously deliver this much-needed data, which the automaker will use to make improvements to its vehicle safety systems and autonomous driving software.

"With help from real-life data we can speed up our development processes and go from years to days," said Ödgärd Andersson, CEO at Zenseact, Volvo Cars' autonomous driving software arm. "As real-time collection generates a lot more data, we can create better and higher-quality data sets that allow us to make better and quicker decisions on the next advancements in safety. We're taking a giant leap to increase safety in and around our cars."

Zenseact is Volvo's technology partner and develops the software powering its vehicles.

The first Volvo model to share its data with the automaker will be the company's new flagship SUV, the Concept Recharge, which will be built on a completely new EV platform.

The core computing system is made up of three main computers. These support each other in operating computer vision processing and AI, general computing and the vehicle's infotainment.

Volvo Cars' upcoming fully electric flagship SUV will have industry-leading safety technology as standard, It will come with state-of-the-art perception sensors, including lidar developed by Volvo's partner Luminar, and an autonomous driving computer powered by the NVIDIA DRIVE Orin system-on-a-chip, as standard. Nvidia's Orin processor is one of the world's most powerful SoCs for autonomous driving.

To process the massive amount of real-time traffic data collected from Volvo vehicles, Volvo Cars and Zenseact are investing in a massive AI-powered data factory within the next few years. The data center can store over 200 Pebibytes (225 million gigabytes) of data. 

By using the power of AI and a NVIDIA DRIVE Orin processor installed in a vehicle, the massive amounts of data generated can be crunched at record speeds for validation.

"To really have a good computer vision system, you need good computing power," said Håkan Samuelsson, CEO of Volvo Cars. "When it comes to the whole experience in the car, which we rely on to differentiate our vehicles, with centralized compute, you can develop that much more flexibly and faster."

The use of real-time data is part of Volvo Cars' long term goal of a future where accidents won't ever happen. Volvo will be able to continuously improve safety and its autonomous driving capabilities to reduce fatalities and accidents as a whole.

"Safety is part of our heritage and the backbone of our company, but software is a crucial part of our modern-day DNA," said Mats Moberg, head of R&D at Volvo Cars. "So while we continue to build on the 50-year expertise of the industry-leading Volvo Cars Accident Research Team, we can now also leverage AI as a new, virtual accident research team."

Over time the car will improve and have the hardware and software capabilities to allow the car to take over steering or braking on its own, in case the driver does not respond in life-threatening situations after repeated warnings, Volvo said. 

In the future, the Volvo vehicle itself will be watching out for human occupants at all times.


Prev                  Next
Writer's other posts
Comments:
    Related Content