For decades we’ve been teased about the possibility of self-driving vehicles. Now we’re finally seeing the first successful autonomous vehicles (AVs) speeding down the roadways. How has it become possible for AVs to be part of our present rather than a theoretical future? Edge computing.
What is edge computing?
When it all boils down, cloud computing is the action of using another person’s computer (or hundreds of computers in a data center) to access your information via the internet. That’s all well and good until the information you need to access is what’s controlling a vehicle hurtling down the interstate that relies on internet connectivity.
Most people have experienced lag rage—like when Netflix takes too long to buffer or your cat picture doesn’t load immediately—but having lag occur for an autonomous car moving at 60+ miles per hour could be a death sentence. As AVs are actually rolling out on the road, what can be done to help ensure they have all of the cloud computing power they need to keep passengers safe? Edge computing is the answer.
Not familiar with edge computing? Edge computing is a type of cloud computing; however, rather than having data produced in a centralized warehouse, like with traditional cloud computing, edge computing occurs when data is processed as close to the edge of the network where the data is generated. The purpose of edge computing is to allow for real-time data processing without having to deal with latency. Moreover, it allows for smart applications and devices to instantaneously respond to data as said data is created.
Edge computing and AVs
There are several aspects of edge computing that make it the perfect vector for making AVs possible.
Communication power
As AVs move along the roadway, they will send out information about the weather, road conditions and any other data their sensors can collect. Other vehicles on the road can then utilize this information to adjust accordingly for any problems on their current route. By being able to transmit data between vehicles rather than having to interface with the cloud, the data arrives quicker and is much more likely to be accurate.
Data management and storage
AVs create a staggering amount of data per day. To set the stage, the 336 million daily Twitter users generate more than 300 GB of data. By comparison, a single autonomous car generates 30 terabytes a day. This is 102.4 times more data! The current network architecture is not capable of handling this much data.
Luckily, with edge computing, it’s possible to determine what data can stay at the edge to be processed by a vehicle’s computer and what actually needs to be sent to the cloud for analysis.
Roadway safety
The rise of AVs could result in the roadways being safer for both drivers and pedestrians. Before this can happen, however, the technology to provide real-time road conditions needs to be perfected.
Utilizing edge computing will allow AVs to gain situational awareness, but this will still need to be processed at the edge and combined with AI/machine learning to fully achieve the level of safety desired. This is especially true since any degree of lag could prove fatal. When life is on the line, AVs can’t wait for cloud processing—even if it can happen in a little as 100 milliseconds.
Moreover, true on the road safety and efficiency won’t be capable until AVs are capable of human-like thinking. This is possible by combining edge data collection with AI and machine learning. As these technologies are being perfected, more car manufacturers are getting on board with the possibilities of AVs.