In 2017, artificial intelligence (AI) and the Internet of Things (IoT) were the hottest technology trends, garnering a lot of attention from technophiles. There is no doubt that these two technologies are going to revolutionize the digital world in the same way smartphones transformed the first decade of the 21st century.
Before delving into the technical details of AI and IoT, let’s start with the basics.
IoT is a set of technologies that manage how physical devices transfer and receive data over the network with no human-to-human or human-to-computer involvement. AI is an area of computer science that creates computers that mimic humans in areas such as speech recognition, decision-making, reasoning, learning, problem solving and translation.
Connecting the dots, IoT gathers vast amounts of data from various environments while AI uses data science to transform the collected data into meaningful information and applies analytics to the data to make decisions. (See Figure 1.) So, collecting data from IoT devices and processing it with intelligent computers to make sense out of these huge data demands new ways of innovating with AI. This combination is expected to enhance and improve the current technology ecosystem.
Figure 1. The combination of AI and IoT results in intelligent machine-based decision making
How IoT benefits from AI
IoT will produce a tsunami of big data as the rapidly growing number of devices and sensors get connected to the network. According to one source, 50 billion devices will be connected to the internet by 2020. The sheer volume of data being created by these devices is expected to grow dramatically for the next few years. This data will be extremely valuable for determining what’s working well and what’s not, such as the collection and sharing of data between self-driving cars.
However, data by itself does not provide value unless it is transformed into actionable, contextualized insight. Many organizations struggle to make sense of the enormous amount of data as it is next to impossible for humans to analyze it in a timely fashion. Therefore, it is imperative that we find a way to transform big data into useful information within a reasonable time period.
This is where artificial intelligence plays an important role in aggregating the flood of data and providing the analytics required to extract meaning from the data. When we feed a haystack of data from IoT devices into an AI system, it reviews and analyzes the data to reveal patterns and similarities that lead to more informed decisions made either by machines or humans.
Here are a few examples of the value that AI brings to IoT applications:
■ IoT can scale. IoT is all about connected devices that share data amongst themselves. These IoT devices range from high-end computers and mobile devices to low-end sensors. There are ecosystems of devices of all kinds but generally, the most commonly IoT ecosystems are comprised of low-end sensors.
By adding AI to the IoT ecosystem, we can make sense of the data. AI extracts information from one device, then analyzes and summarizes it before transmitting it to other devices. This reduces the flood of data to a manageable level and enables a larger number of IoT devices to be connected to the network. This is called scalability.
■ IoT becomes smart. In IoT applications, AI-based machine learning is used to a limited extent to react to unexpected situations. When a device receives an unexpected query or detects unusual conditions, it needs to know whether to respond autonomously or sound an alarm for human assistance. To make such a decision requires intelligent learning and decision-making capability. Google uses this approach in the RankBrain algorithm, which uses deep learning to guess the meaning of an unexpected query, then responds in real-time without any human intervention.
■ IoT boosts efficiency. With the help of predictive analytics, machine learning coupled with AI learns from the data to decipher trends and make predictions about future events, such as the likelihood of the failure of a piece of factory equipment so that it can order replacement parts just in time. This unlocks the real benefits of IoT in a variety of manufacturing industries.
■ IoT saves lives. In the healthcare sector, the addition of AI to the IoT network can dramatically improve patient care. Health-oriented smart devices are used to track physical activity, heart rate, body mass, temperature and other vital metrics. AI sifts through the data to identify individual behaviors that are outside the norm and might indicate a health problem. AI will then analyze the data from health devices, review the user’s medical history and if necessary alert the patient’s doctor accordingly.
These are just a few promising applications of artificial intelligence in IoT. Also, AI can help us create systems that learn from the data they process.
The challenges ahead
■ Security. Since AI systems and IoT networks are constantly collecting and analyzing sensitive data, it’s important to ensure the data is secure and safe. Confidentiality and integrity of both the data and the systems can be compromised at any time by hackers attacking the ecosystem. A security breach could be very costs for an enterprise by disrupting services, angering customers and damaging the brand.
■ Compatibility and complexity. IoT is a collection of many kinds of devices that use many different types of technologies. This can cause difficulties and require the deployment of hardware and software interfaces when connecting the devices. Large numbers of diverse devices and interfaces lead to more complex ecosystems.
■ Artificial stupidity. It is a derogatory reference to the inability of AI programs to adequately perform basic tasks. AI systems and the algorithms AI uses need to be well developed to understand and interpret data so more accurate and sensible decisions can be made. This is a work in progress.
IoT is “Batman,” AI is “Robin”
IoT is useless without its sidekick AI-based technology. Combining AI technology with IoT allows IoT ecosystems to scale to millions of devices capable of processing huge volumes of data. To be able to find a needle in this haystack requires the teamwork of IoT and AI. In the months and years ahead, AI will be an essential part of IoT systems and will take IoT to the next level.
About the Author
Senior Software Engineer
Raj Kumar is currently working as a Senior Software Engineer at Aricent and has 3+ yrs. of experience in Core Java, JMeter, Html, JSP, and Selenium WebDriver. He holds B.Tech degree from Cochin University and is passionate about new technologies like AI, IOT etc. Raj likes to play chess, listen to music in his spare time.