To Be Autonomous, Cars Need to Be Like Mobile Data Centers
Outfitted with advanced sensors and computers, fully autonomous cars might not even look like the cars of today. They might be more like terabyte-storing, data-crunching mini data centers on wheels.
In the self-driving car of the future, data could very well be digital fuel. There is the vehicle data in the autonomous application stack and operating system. Map data flows from 3-dimensional, high-definition maps and in-vehicle navigation systems. Not to mention, the data being written to and read from:
- Dashboard Cameras
- Digital Clusters
- Drive Recorders
- Telematics Gateways
- V2X Communication Systems
- And more!
With these growing demands, vehicles could soon be faced with storing and processing over 1 terabyte of data. Doing so will require a new approach to data infrastructure. To bring about this change, onboard computing and edge processing are playing a leading role. Only by capturing and analyzing data can autonomous vehicles react to changing road conditions and make split-second driving decisions. To learn more about the future of onboard storage and compute, we spoke with Josh Miller, Senior Manager, Strategy at Porsche Consulting.
“In an autonomous world, the need for increased V2X communication will grow significantly. We’ll have much higher demand for point-to-point communication that doesn’t travel on high-latency back ends,” says Josh Miller, Senior Manager, Strategy at Porsche Consulting.
Addressing Massive Storage Demands of Autonomous Vehicles
Human drivers naturally process tons of data while on the road: weather patterns, road conditions, traffic movement, pedestrians, and more. To mimic this behavior, autonomous vehicles use a variety of tools to “see” the road. These include imaging systems, radar, accelerometers, telemetry, gyroscopes and more. That’s a ton of data being generated, something like 4 terabytes per day based on an average driving time of about 1.5 hours1. Current vehicles are ill-fit to handle storing and processing such massive workloads onboard.
“We’re starting to see vehicles as mobile data platforms. If you look at purely the storage and compute to support fully autonomous vehicles, it’s a step-function difference,” describes Josh Miller. “So, you have in some of the pilot vehicles out on streets today, mobile server farms with high storage and high compute. They are supporting multiple autonomous drivers that are monitoring each other.”
This adds another layer of complexity. Autonomous vehicles must decide which data is absolutely necessary to make driving decisions, and then perform near-instant analysis. Not to be wasted, the unused data could be uploaded to the cloud for OEMs to use in future insights. This vehicle data – whether structured or unstructured – could be gathered from autonomous vehicles around the world and placed in a data lake2. That way, algorithms using artificial intelligence and machine learning could analyze historical data to better understand how autonomous vehicles can make safer driving decisions.
What Will Drivers-Turned-Passengers Do with Their Free Time?
No longer worried about navigating the roads, former drivers will be freed up. One source even estimates 250 million hours of commuting time saved for people in the most congested cities in the world3. This begs the question: What will passengers do with their new found free time?
The answer – well, answers – might center on getting pesky chores done. Imagine a commuter who wants to finish their beauty routine before work. An autonomous vehicle could be outfitted as a mobile beauty salon, complete with hair and nail mini treatment stations. Similarly, turning a driverless car into a small-scale clinic could help passengers carry out health exams in route. Not to forget, taking care of online shopping – whether at local stores, pharmacies, or retailers that could enable passengers to ride for free. The possibilities for mobility services are widespread, but finding the path to profit is a challenge.
“I think we’ll see many different players in the [connected car] space trying to figure out how they can either present content, media or other information to figure out the best path to monetization,” – Josh Miller
Why the Cloud is Not Enough for Self-Driving Vehicles
Remember being taught the importance of clear and consistent communication? Perhaps it was to get a group assignment done in school. Maybe it was to finish a complex project at work with your colleagues. In either case, having open lines of communication helps everyone get on the same page, understand what’s going on, and make informed decisions.
The same idea holds for autonomous vehicles. Each automobile needs to “talk” with the other cars on the road and the greater driving infrastructure to keep passengers safe. But accessing the cloud might run into problems with slow network speeds or spotty connections. Even though embedded flash drives can overcome some of these problems, there is more needed for real-time decision making.
Enter: edge processing. The “edge” is an idea for a network connected by gateways that pull together data from multiple cars on the road to provide helpful services. In the case of autonomous vehicles, these gateways would enable real-time communication between automobiles to navigate safely through congested traffic and other dangerous road conditions. Edge gateways will need to be placed strategically in driving environments and be tough enough to survive temperate and weather extremes.
MaaS (Mobility-as-a-Service) Disruption in Automotive
Turning the driving experience over to autonomous systems wouldn’t only mean more free time. It could also mean more people who forgo owning a car full-time. Instead, travelers might have their primary mode of transportation be through ride hailing services, a convenient and cost-effective alternative. Then, individuals or businesses could lease private vehicles to turn into autonomous taxi fleets, whisking people around the city. If such a worldwide passenger economy were to come to fruition, it could present a nearly $7 trillion total opportunity for consumers and enterprises by 20503.
Before MaaS becomes a household acronym, there are a few steps that would need to happen. One stage is vehicle sharing, which we mentioned previously. Another important factor is the rise of urbanization. Globally, the number of urban dwellers is projected to grow from 55% today to nearly 68% by 20504. With this trend, a new wave of commuters within cities might rise, too. The third measure to be accounted for in mobility-as-a-service is connectivity. After all, an autonomous vehicle is actually an incredibly complex IoT device5. More powerful telematics and V2X communications could make more advanced services for passengers possible.
With autonomous vehicles, passengers could enjoy smoother commutes and a more enjoyable experience on the road.
“We’ll see what emerges with leading service providers and how they get presented,” says Josh Miller about the future of in-vehicle experiences. “It will be an ecosystem that evolves over time.”
Learn more about the storage challenges of edge computing for autonomous vehicles here.
FORWARD-LOOKING STATEMENTS: This article contains forward-looking statements, including statements relating to expectations for storage products, the market for storage products, product development efforts, and the capacities, capabilities and applications of Western Digital products. These forward-looking statements are subject to risks and uncertainties that could cause actual results to differ materially from those expressed in the forward-looking statements, including development challenges or delays, supply chain and logistics issues, changes in markets, demand, global economic conditions and other risks and uncertainties listed in Western Digital Corporation’s most recent quarterly and annual reports filed with the Securities and Exchange Commission, to which your attention is directed. Readers are cautioned not to place undue reliance on these forward-looking statements and we undertake no obligation to update these forward-looking statements to reflect subsequent events or circumstances.
- For Self-Driving Cars, There’s Big Meaning Behind One Big Number: 4 Terabytes. https://newsroom.intel.com/editorials/self-driving-cars-big-meaning-behind-one-number-4-terabytes/#gs.fx32xq
- Autonomous Cars, Big Data, and Edge Computing: What You Need to Know. https://dzone.com/articles/autonomous-cars-big-data-and-edge-computing-what-y
- Welcome to the Passenger Economy. https://www.intel.com/content/www/us/en/automotive/passenger-economy-report-infographic.html
- 68% of the world population projected to live in urban areas by 2050, says UN. https://www.un.org/development/desa/en/news/population/2018-revision-of-world-urbanization-prospects.html
- How IoT is Driving the Autonomous Vehicle Revolution. https://www.iotforall.com/iot-and-autonomous-vehicles/