Oct. 31, 2021
Edge AI is everywhere—and if you haven’t noticed it already, you certainly will very soon. The intersection of artificial intelligence and edge computing, edge AI prefers to perform computations on “edge” devices that are physically located close to the source of the data.
Perhaps the most familiar (if still elusive) example of edge AI is the self-driving car. Sensors and cameras in the car are constantly collecting data, images, and videos that need to be rapidly analyzed in order to steer the car, control its velocity, and avoid obstacles and other drivers.
Although this data could theoretically be uploaded to the cloud for processing and analysis, this presents a few issues:
Instead of performing computations in the cloud, self-driving cars have edge AI devices located within the car itself that are powerful enough to analyze the camera and sensor data.
Edge AI examples and use cases can be found across dozens of industries. For example, edge AI is helping to transform the manufacturing sector. Cameras, sensors, and other devices are constantly collecting vast quantities of data, which is then analyzed at the point of collection by edge AI. The use cases of edge AI for manufacturing include monitoring product quality and rapidly detecting issues with manufacturing equipment, thereby helping to improve worker safety and reduce the costs of defects.
Another tremendously popular use case for edge AI is retail. Large retail stores use real-time video analytics for a variety of purposes: counting the number of customers in the store throughout the day, calculating how much time customers spend in each area of the store, etc. Instead of sending the images and videos to the cloud, all of these computations can be performed on edge AI devices that are physically located inside the store.