How AI Plays a Role in the Internet of Things
The growth of the "Internet of Things" has exploded since the term was coined in 1999 and the first Internet-connected smart refrigerator announced a year later. Now, applications of IoT are everywhere. In the home, in recreation, in healthcare, on the road in cars and trucks, on the rails, at sea and in the air, in warehousing and logistics, in retail and financial services, in manufacturing and process control, in agriculture and farming, and in the military.
In a previous post on the Internet of Things (IoT), we looked at examples of IoT that many readers would encounter in their daily life. We looked at applications in home heating and cooling, in the kitchen and home security, on-the-road, and in healthcare. In this post, we'll look at how artificial intelligence (AI) is being used alongside the sensor and connected device technologies that are at the heart of IoT.
How to Prepare for a Career in AI and IoT
Earning cloud certifications is currently the best way to learn the skills needed to work with AI and the IoT. Just about every IoT application or service relies on the cloud, which also means you'll need a strong grasp of networking and security topics.
A Look at Artificial Intelligence
It should be no surprise that surveys of IT executives show that combining AI with IoT is a top priority initiative for many organizations. They see that AI can facilitate automation of tasks that would normally require the involvement of a person, such as speech recognition, speech-to-text, facial and image recognition, language translation, analysis and decision making. Before we explore specific examples of the way in which AI is being used in IoT implementations, let's consider why these technologies are so complementary!
A key aspect of IoT applications is that they collect massive amounts of data. These data stores can then be processed in real-time, or analyzed after-the-fact, by software to derive actionable information.
An example is the car navigation app that acts in real-time on vehicle positional data to compute traffic speeds, deduce where there are slowdowns and jams, and offer alternative routes. In a manufacturing example, an app can analyze production line performance data and project potential points of breakdown. These are both examples of artificial intelligence being used in conjunction with connected IoT sensors and devices. In both cases, the applications are following paths coded by their developers. A subset of artificial intelligence is machine-learning, where the app learns from the data and refines its decision-making without explicit coding.
The Benefits of AI + IoT
There are many business reasons for organizations to apply AI technologies to their IoT applications. For example, combining AI with IoT allows organizations to:
Improve their operational efficiency: For example, AI-enabled applications can make data-driven decisions and logic choices in real-time, helping boost the efficiency of business operations. Likewise, analysis of streaming sensor data for predictive maintenance can help avoid the expense and disruption of breakdowns and related unplanned production outages.
Help reduce enterprise risk: Risk comes in many shapes and sizes—physical and cyber theft of assets and intellectual property, organization exposure to financial loss, employee injury, product liability, etc. AI can be used in conjunction with IoT applications to help monitor and warn against known indicators of risk. For example, proximity sensors and automatic machine shutoffs are just one way to protect employees from serious injury and the enterprise from financial risk.
Expand the scale of their operations: Using AI to monitor, analyze, and process data streams from network-attached sensors and devices means that organizations can handle much larger IoT networks.
Introduce new and improved products and services: Companies can use AI to add 'smarts' to new and existing products and services. Using the inherent flexibility of software, they can tap into the data generated by IoT devices, provide powerful user interfaces, and much more!
How AI and IoT Work Together
Let's take a look at some specific examples of how artificial intelligence is used to add value to the Internet of Things applications.
AI and IOT are used extensively in both discreet and process manufacturing sectors. Whether it's an assembly line robot, a process control machine mixing chemicals, or making pharmaceuticals, manufacturers are using smart AI-driven devices and sensors to reduce costs and improve production efficiency.
We've already touched on predictive maintenance! Here, a smart system collects and analyzes sensor data to assess the performance of individual machines, as well as the overall end-to-end production line. By comparing actual with projected performance, these systems can pinpoint potential points-of-failure and recommend preemptive maintenance.
Although the headlines are captured by the likes of Tesla and Google's Waymo with their self-driving vehicles, there are plenty of examples of AI working with IoT in your car or truck. For example, your car is probably equipped with Lidar sensors, which enable smart applications such as lane departure warning and lane keep assist, adaptive cruise control (ACC), antilock braking (ABS), collision avoidance (CAS), and traction control. With a lane departure warning system, the combination of sensors and AI alerts the driver that they are straying out of the lane, while lane assist nudges the steering wheel to take them back into the lane.
Adaptive cruise control maintains a set speed and—using front-facing radar—can detect slower vehicles ahead and will automatically slow down and maintain relative position. Using the same sensors, the collision avoidance systems will automatically apply the brakes if it determines that a collision is imminent.
New AI-enabled versions of intelligent cruise control will leverage GPS mapping technology to allow speed to be adjusted as the vehicle approaches road junctions, traffic lights, curves, and turns.
Speaking of GPS applications. There are numerous examples where AI and IoT combine with GPS for great effect. Think for example of popular navigation apps like Waze and Google Maps. Using a GPS satellite connection to fix the vehicle's location and cellular service to transmit and receive data, these apps can in real-time detect a car's changing position on a specific roadway. The apps are community-driven, which means that they capture positional data from other users on the road. By comparing point-to-point positional data, the apps can compute the speed of your vehicle and — combining that with the positions of other cars on the road — deduce traffic slowdowns and jams. The navigation app can also detect where on the road cars resume normal speeds and, using this information, it can tell drivers how long they can expect to be in traffic, as well as offer routing alternatives.
One of the first widely used portable navigation devices for cars was produced by Garmin. This company now provides GPS-enabled IoT products for cars, boats, and planes, as well as wearables such as fitness trackers and golf watches. Smart golf watches are equipped with GPS, gyroscope, and accelerometer sensors, as well Bluetooth® Smart connectivity. The watches use GPS to determine which golf course you are on and exactly where you are on the course. By accessing a downloaded 'map' of the course, the watch's software can tell you how far it is to the green and to various intervening hazards—bunkers, water, doglegs, etc.
The watch can also detect when you hit the ball and can compute how far you hit it. In an example of machine learning, if you tell the watch which club you used for each shot, then it will begin to learn your range of shots, and like a human golf caddy, can recommend which club you should use for your next shot.
Healthcare and Wellness
Healthcare and wellness provide another rich vein of applications of artificial intelligence and IoT. Fitness trackers are the most common example. They come in many forms — rings, bracelets, watches, and pendants — and have sensors such as accelerometers, gyroscopes, and altimeters, as well as GPS and Bluetooth connectivity. Typically, fitness trackers will come with the AI 'smarts' to count steps taken and compute the distance traveled and speed. They may also allow you to set up and track a tailored workout program, including heart rate and more. Some advanced ones will learn from indicators such as your body temperature and modify your workout accordingly.
The healthcare sector has adopted IoT and artificial intelligence to radically improve the level and timeliness of care that can be offered to patients. Examples of so-called Internet of Medical Things (IoMT) devices include smart beds, telemetry devices for intensive care, IV pumps, and ventilators.
The combination of IoMT and artificial intelligence is being used to help diagnose, predict future illnesses, manage the administration and effects of medications, and provide ongoing treatment of chronic conditions.
Common applications of IoMT are for patient monitors and trackers, which can be used for diagnostic purposes, for patient surveillance, or even medical intervention. Devices such as infusion and insulin pumps can be used by patients with chronic conditions in the healthcare facility, or in their homes.
There have been significant advances in sensor technology. Sensors can measure vital signs such as heart rate, temperature, blood pressure, oxygen and glucose levels. These devices can be attached by patches, surgically implanted, wearables like a watch or as smart clothing, or even ingestible. One recent startup developed a rice grain-sized sensor to be ingested with medication. On contact with the patient's gastric juices, the sensor transmits the type of medication, dosage, and time to a patch worn by the patient. From there, the data can be transmitted to the healthcare provider's portal for nurse or doctor review.
And you don't need medical-grade devices. In a study of healthcare workers wearing off-the-shelf Apple Watches, researchers detected heart rate variability (HRV) that signaled the onset of COVID-19 up to seven days before the infection was diagnosed with a nasal swab test.
We have covered a selection of AI and IoT applications in manufacturing, automobiles, sports, and healthcare. There are examples in many other sectors like retail with cashier-less supermarkets and, of course, smart homes and kitchens. Suffice to say that the Internet of Things and artificial intelligence will provide rich opportunity for IT professional advancement. You can be sure that AI platforms on public cloud services like AWS, Azure, and Google will be at the center of this opportunity!
If IoT and AI are on your career trajectory, then we suggest that you should consider cloud certifications starting with AWS certification training options for Certified Cloud Practitioner, Certified SysOps Admin, or Certified Security – Specialty. If your organization is committed to Microsoft Azure or the Google Cloud platform, then take a look at certification training for the Azure Administrator, Azure Security Engineer Associate, or the Google Associate Cloud Engineer.
delivered to your inbox.