You might not need to get behind the wheel of your car in 20 years, but big data will. As we move towards an autonomous future for the automobile, human input will gradually be replaced by computer systems. At the heart of these systems will be massive amounts of data, constantly analyzing, estimating and evaluating the current situation on the road and forecasting future risks.
Self-driving car technology must tackle the three Ps: perception, prediction and planning. The biggest obstacle remains the prediction. This is not necessarily what other autonomous vehicles will do, but rather what other human – or animal – road users will do. One way to solve this problem is deep learning – or pairing of high performance models.
All of this processing means a mountain of data, transferred from every car. If you consider that microchip shortages are causing a supply crunch in the auto industry right now, think about what’s to come.
It doesn’t take a computer science degree to realize that managing, processing and storing data is going to become a major global issue as we move towards self-driving vehicles.
According to Jeff Fochtman, senior vice president of data storage giant Seagate, “Storage is definitely a bottleneck. As the IoT comes online and more information is created by sensors, the severity of this data is overwhelming.
“I really think of technology as three real components: compute, networking and storage, and they all have to evolve together. “
Seagate has developed a new on-board storage device, Lyve, the size of a shoebox installed in autonomous test vehicles. It manages the data flow of these vehicles, which can reach 150 terabytes (TB) per day. To put it in perspective, that’s the equivalent of 32,000 DVDs. Or to put it another way, the entire US Library of Congress was estimated to contain around 74TB of data in 2009.
According to Fochtman, during the self-driving car testing phase, the goal is to collect as much data as possible and not delete it or make a decision. “It’s about collecting data, ingesting that data, creating clouds to build the neural network on which AI works. “
He estimates that we will be in this testing phase for the next 10 to 15 years. “Even if the autonomy levels are reached, the test phase which only continues to collect all the data. Sometimes there are 30-40 sensors, including seven cameras, three LIDAR units. It’s just a matter of spinning so much data and nobody wants to throw anything away as it accumulates. Data is the competitive advantage as we move into regulatory modes. “
Real world data
This is why Seagate is a key partner of the newest Future Mobility Campus Ireland (FMCI). Based in the Shannon Free Zone in County Clare, it offers companies like Seagate, Cisco, Red Hat and Jaguar Land Rover the ability to test autonomous vehicles, both on the ground and in the air, under conditions real.
The test facility consists of a road network that is modernized with advanced inter-connected sensing and telecommunications technologies. Jaguar has already tested its autonomous I-Pace outside of the facility.
Previously, such tests were performed at facilities in sparse, sunny climates like Arizona in the United States, but pitting these vehicles against Irish weather and road conditions provides much more real-world data, according to speakers at a recent FMCI congress.
All of this data is fueling the development of programs to address the scenarios facing future autonomous vehicles as we move towards transferring control from our cars to computers. Before that, we need to deal with the way it’s collected and find a way to create more chips to keep regular cars on the road, let alone future self-driving cars.