The next evolution of artificial intelligence is being driven by a hybrid cloud model that uses cloud and edge resources selectively to maximize the value of artificial intelligence while maintaining latency constraints, costs, security and privacy.
In everyday parlance, the term “on the edge” generally refers to any extremes, but in the context of artificial intelligence and edge computing, “artificial intelligence on the edge” is becoming more and more normal than in exceptional cases. Attention in part due to recent advances in edge computing power and increasingly lightweight AI algorithms, but the real need for edge AI has arisen from the changing dynamics of human-machine interactions and models of user participation.
“Artificial Intelligence on the Edge” is a combination of two key technology areas: Artificial Intelligence and Edge Computing. Artificial intelligence or artificial intelligence is now a common technology in applications such as computer vision, language processing and complex data analysis. AI applications have traditionally been implemented on cloud servers with data transferred to and from the source. For example, video from a surveillance camera can be sent to a cloud, where it is processed by an AI model to detect abnormal behavior.
On the other hand, edge computing enables data processing at the edge of a network closer to the data source through edge software and hardware components: cell phones, IoT devices, smart appliances and processors in an autonomous car are all edge devices. AI on edge refers to AI algorithms implemented on edge devices rather than in the cloud.
Artificial intelligence as a technology began to gain ground with the increasing ubiquity of GPUs (Graphics Processing Units), which are well suited to complex models of artificial intelligence. The emerging demands on consumer and industrial applications reveal some of the shortcomings of centralized cloud computing and advocate state-of-the-art solutions.
Data security and data protection: With increasing amounts of data generated by sensors and devices, the protection of personal data and the security of confidential company information are critical requirements that cannot always be guaranteed with applications in the cloud.
Latency: With cloud-based AI, transmission of data back and forth from the cloud and can negatively impact latency and user experience.
Bandwidth and Cost- Cloud-based models that process large amounts of data can be impacted by bandwidth constraints and the cost of moving to and from the cloud .
Customization and personalization – There is a growing need for customization for the consumption of industrial and business applications like as well as machines. Centrally managed cloud solutions are cumbersome to adapt and maintain.
Real-time decision making: Centralized cloud solutions with higher latencies are inefficient for applications that require automated real-time decision making.
Edge devices with dedicated AIPs (AI processors) or NPUs (Neural Processing Units) now have increasing computing power available that can efficiently and effectively execute complex AI algorithms. The uses for
edge AI are vast and diverse, but here are some industries that seem poised for transformation.
Manufacturing: With increasing automation in manufacturing, there is a need for automated real-time process control with inline sensor data. Edge AI based real time closed loop control can deliver higher efficiency, lower costs and improved quality for manufacturing.
Automotive: Advanced Driver Assistance Systems (ADAS), which represent a step towards fully autonomous cars, are intended to reduce accidents and ensure safety. Given the low latency of Edge AI solutions, they are at the heart of driver assistance applications that demand fast response times for driver feedback and control.
Media and Entertainment: Entertainment is becoming increasingly personalized and personalized with the popularity of social media platforms, over the top (OTT) content, and games. AI solutions on users’ peripherals can enable personalization of the user experience based on their preferences and context without compromising privacy.
Healthcare: With the recent Covid19 pandemic uncovering cracks in the tanks of global healthcare systems, there is an urgent need to efficiently prioritize resources while maintaining high standards of care. When deployed on wearable devices or edge hardware in healthcare centres, Edge AI can help detect and diagnose disease while providing insight and advice to healthcare providers.
While
Edge AI will undoubtedly be transformative for many industries, cloud-based solutions will continue to serve applications that contain large amounts of data or require centralized processing and decision-making. The next evolution in AI technology will be driven by a cloud-edge hybrid model making selective use of both in the cloud and on edge resources to maximize the value AI delivers while meeting constraints on latency, cost, security and privacy.
