What is edge computing? Well, it will likely be the key to a global Industrial Internet of Things (IoT) infrastructure (especially autonomous vehicles). Unlike cloud computing, edge computing pushes storage and processing closer to the source of data and not to a large centralized system. In other words, edge computing optimizes smart devices to store and process information. Thus, it distributes computing power where it makes the most sense (e.g. at the “edge”).
There’s no better use case for edge computing than supporting the evolving autonomous and connected vehicle infrastructure. Why? Because it enables data to be processed closer to where it’s created (i.e. motors, pumps, generators, or other sensors). In turn, this reduces the need to transfer data back and forth, to and from the cloud.
In terms of communication, each autonomous vehicle (AV) broadcasts information on road conditions, weather and other sensor-based information. That information can be used by other vehicles to adjust for detours, debris, accidents, etc. Instead of interfacing with cloud servers, much of the data can be sent and received between vehicles. This allows for quicker and more accurate utilization. Edge data centers will become essential for the offloading of non-critical data and keep more time-sensitive, critical data within the vehicle.
Did you know? 336 million Twitter users produce over 300 GB of data every day. That’s quite impressive until you compare it to one autonomous vehicle which generates a staggering 30 terabytes per day (or about 3,500 4k movies)! There’s no way the current network architecture could handle that kind of volume. Especially if you consider the fact that there will be 10 million self-driving cars on the road by 2020.
At the end of the day, edge computing will determine how much data needs to stay on the edge and processed by the vehicle’s onboard computer versus what needs to be sent back to a centralized cloud for analysis.
Autonomous vehicles might be a solution for making our roads safer for drivers and pedestrians alike, but they need to see road-conditions in real-time. Edge computing helps AVs achieve situational awareness by combining information collected and processed at the edge and applying AI/machine learning. Even a millisecond lag can be the difference between life and death. AVs can’t afford to wait for information stored and processed in the cloud, even if that transaction only takes 100 milliseconds, three times faster than a blink of the eye.
There’s more to it, though. Data collected at the edge needs to be combined with AI and machine learning for self-driving cars to think like humans and drive more safely and efficiently. Major car manufacturers are starting to buy into autonomous driving, but further progress in safety is needed. At Motus, we’re constantly thinking of ways to make life on the road safer for the mobile workforce.
Edge computing is most suitable for bandwidth intensive and latency sensitive tasks (like sending 30 terabytes of data or transactions that take less than 100 milliseconds). But there’s still a role for cloud computing. Until Smart Cities and their infrastructures are firmly in place, a combination of cloud and edge computing will be the best approach. Using the strengths of each technology will be beneficial in the short term. Edge computing can prepare an instant response to localized data and transmit it – when ready – to the cloud for more complex processing. Want to learn more? Subscribe to the Motus blog for more posts on the future of automobile technology as a whole.