The Internet of Things is becoming increasingly important both in the consumer as well as industrial markets. For the latter, a common characterization is Industrial Internet of Things (IIoT). We might also find references to Consumer Internet of Things (CIoT). The availability of high bandwidth Internet connections especially with the emergence of robust 5G networks is giving rise to innovative IoT solutions in Smart Cities, Automotive, Industry 4.0, Supply Chain, Healthcare, and Energy. In our homes, businesses and enterprises we are surrounded with connected devices.
According to IDC there will be more than 40 billion devices generating close to 80 zettabytes of data by 2025. Other predictions put the number much higher. “Things” come in different sizes and shapes — from small sensors to large turbo engines. In the consumer market, Smart Home devices and various home monitoring, smart appliances, smart speakers, smart entertainment and energy efficiency solutions are becoming pervasive. The projected total IoT market is estimated to be in the trillions.
Not surprisingly Artificial Intelligence has also emerged as an increasingly important enabler of connected Things: from Natural Language Processing in smart speakers to predictive maintenance of connected devices to AI applied in a plethora of advanced deep learning applications. A perfect storm of technological advances in computing, databases, and networking has caused the emergence of AI as one of the most significant digital transformation trends. Another trend of course is IoT.
So what is the value proposition of AI to IoT? Interestingly there has been some coverage of Intelligence of Things (the other “IoT”) — as covered in CES 2020 this year — or more explicitly Artificial Intelligence of Things.
The various connected devices generating data for various sensors such as air pressure, temperature, motion, and many other parameters provide opportunities to analyze, mine and discover AI models. These models can then be deployed for application areas such as predictive maintenance.
When it comes to autonomic IoT, the emerging connected devices trend is Edge Computing. The Industrial Internet Consortium defines it as follows: “Edge computing is a decentralized computing infrastructure in which computing resources and application services can be distributed along the communication path from the data source to the cloud.” Edge computing allows devices to take immediate action. In other words, edge computing is faster and more optimized.
Increasingly business logic and AI models are executing at the IoT edges. They are many advantages to executing AI models at the edge. In many applications where instantaneous decisions need to be made dealing with, for instance, hazardous material, or dangerous levels of pollution or even emergency dispatching the delays of round-tripping from a Cloud-based data center for AI model execution could be prohibitive. In some cases, Internet connectivity might have unacceptable latencies or even be off-line. The actuators and sensors of connected devices often leverage gateways to connect to the Cloud. The connected devices as well as the gateways are becoming increasingly powerful in storage, computation, and networking optimizations.
Pushing computations and intelligence to the edges reduces the latency of the decision executions. This could be critical in many industrial and consumer IoT applications.
Formula 1 is an interesting use-case of AI and edge computing. Race cars that aggregate 100 Gigabytes of data per car over a race weekend. There will be “more than 100,000 data points are streaming from a single McLaren track car per second.” Tire change, safety, and gear change analysis are among some of the decisions that need to be made in real-time by the engineers at the trackside or mission control. Intelligence needs to be pushed to the edges and acted upon in the speed of the race!
Another, more current example in this Covid-19 era is Intelligent IoT Edge Computing for Medical Things. Due to the lock-down, connected monitoring, connected tests, and telecare are providing tremendous opportunities for remote patient monitoring in these difficult times. The pandemic has stressed the care providers and hospitals — especially when the number of patients increases sometimes exponentially. Remote monitoring of connected devices and the intelligence pushed to the edges of the monitoring devices becomes critical in alleviating some of the pressure from the caregivers and even saving lives. For instance, recently GE Healthcare introduced Mural Virtual Care Solution, which aggregates data and monitoring from several systems and allows virtual monitoring through “near real-time data from ventilators, patient monitoring systems, electronic medical records, labs, and other systems.” The solution “allows one clinician to monitor several patients at once, supplementing existing monitoring devices in patients’ rooms.” At its core, it is an intelligent virtual ICU system with tremendous benefits both for the caregivers and patients — monitoring the progress of the patients across geographical distances.
In conclusion, the connectivity of devices in our homes, hospitals, cities, and industries is becoming ubiquitous. A number of technologies such as 5G, Cloud, and AI are critical for robust and pragmatic solutions with IoT. Connectivity is exploding but more importantly, edge computing is becoming critical. It realizes the benefits of IoT in real-time. The faster intelligence is applied where the devices and people are, the better. The edge intelligence and its application is a critical trend that will accelerate in the coming years.
