Artificial Intelligence of Things (AIoT)- What is artificial intelligence of objects? In this article where we will try to answer the question, I wanted to give an IoT date before.
The concept of the Internet of Things is 22 years old. In 1999, Kevin Ashton, who worked in procter&gamble supply management department, used the concept of internet of things as the first term when describing the application developed to increase efficiency with radio frequency technologies. But no one, including Kevin, showed any value for that term for about 10 years. However, the history of internet of things application examples, which means that objects communicate with each other, date back to the 70s. However, before that, the terms ’embedded internet’ or ‘common calculation’ were used.
The concept of IoT, which everyone adopts regardless of the Internet of Things or the mother tongue, began to gain some popularity in the second half of 2010. During the same period, google’s StreetView service did not only take 360-degree photos, but also kept tons of data about people’s wifi networks. With this agenda, it became clear that Google not only had internet world data, but also a strategy related to the data of the physical world.
Coincidentally, in the same year, the Chinese government announced in its Five-Year Plans that it would make the Internet of Things a strategic priority.
By 2011, Gartner, the market research company that invented the famous “hype cycle for emerging technologies,” included a new phenomenon on their list: the “Internet of Things”.
In the following years, conferences, fairs and articles themed on the ‘internet of things’ followed each other.
The term Internet of Things reached mass market awareness in January 2014 when Google announced its acquisition of Nest for $3.2 billion. The Consumer Electronics Show (CES) in Las Vegas was held under the theme IoT.
The term was originally used for an RFID project, but it contains a much larger vision than this project. The Internet of Things is the interaction of the physical world with applications with business intelligence on central platforms, equipped with sensors. We may think of m2m that almost all of us are familiar with, but IoT includes a different and more inclusive vision than M2M. The ‘measure-watch’ motto that came into our lives with M2M has been developed as a ‘measure-watch-compare-increase-understand’ in IoT. In other words, the concept of the Internet of Things targets interdisciplinary exchange of information.
Over the past decades, the concept of the Internet of Things has been discussed many times and its benefit and applicability have been questioned. With 5G, the proliferation of IoT applications can be discussed, while we’ll talk about why global players care about the concept of Artificial Intelligence of Things (AIoT).
5G and IoT
5G and Features; It is included in the Business Plan of Institutions
Although high-speed broadband tops the list as the most demanded, we can say that demand has decreased compared to last year, but overall demand in 5G has increased by one point. Despite all this, the institutions that will invest in IoT; It is very difficult to know what they prioritize when it comes to 5G and its features. But whether institutions really need all the advanced capabilities of 5G is currently a secondary question; because interest in 5G will bring new collaborations, together new opportunities for value creation and innovation.
GSMA Intelligence; It is clear that the majority of respondents are familiar with the capabilities of 5G and consider them important to allow successful IoT implementation; however, he notes that they do not yet fully understand how different abilities will benefit them privately. (businesses seem to want it all). 5G’s improvements in terms of lower latency, higher transmission speeds, and increased network capacity (large IoT) should be seen as opening the door to enterprise digital transformation.
Which 5G Feature Is Critical for IoT Applications?
|5G Feature||Very Important||Important||Junk||I do not know|
|URLLC (Low Latency)||%52||%37||%8||%3|
|MMTC (large IoT)||%49||%39||%8||%4|
Source: Enterprises speak: IoT gets real
Enterprise LTE (Corporate Networks – Private Network) – Demands and Requirements
In the GSMA Intelligence Report, Corporate LTE also included corporate 5G (Enterprise Network-Network) or Private Network requests in its survey, which is still offered today but is considered one of the strongest value offerings of 5G.
Demand for corporate networks has not changed. Almost a quarter (22%) of businesses surveyed said they needed location-specific coverage in 2020. This is the same as in 2019, which is interesting, especially given recent developments and announcements in the automotive and public sector. We can understand this as a decrease in interest in location-specific coverage and a decrease in demand for enterprise networks.
Hype factor effect. When asked about the importance of private networks for successful IoT projects, 55% of businesses said they were very important. When the same question is asked “is private, localized coverage better than that available on existing public networks (covering your facility or campus)”; this figure, dedicated to the importance of private networks, fell to 48%. Considering that 22% of businesses really need private networks; it shows that perceived importance does not turn into real business needs or requirements.
Edge Computing: Role and Importance in IoT
There’s a slight information gap. 46% of businesses think edge information is very important; however, when asked to process large amounts of data on-house or at the edge (e.g. device, gateway) to improve efficiency, 51% say it is very important. This points to a small gap of information about what end unit analys is really.
Are we really talking about the end? Most IoT traders believe that IoT projects already use edge unit analytics, and this is; it may seem wrong given the imlessness of the technology. Also, not all IoT traders – decision makers know where the data is stored and/or analyzed; 11% of those claiming to know say the data is stored at the edge, while 36% say the data is analyzed at the end (on the IoT device/gateway/modem).
Summary of 5g and IoT Relationship
In the section so far, based on the GSMA Intelligence Enterprises speak: IoT gets real report, I tried to draw attention to the small information gap that shows the relationship between 5G and IoT and the inconsistency of the answers given by investors when it comes to the internet of things. Although 5G and IoT are interconnected topics, it is unfortunately still very difficult to see the bigger picture in both. The main reason behind this situation is that the ‘Killer Application’ scenario (in foreign terms) has not been clarified. What is the most common and basic value offer of 2G,3G and 4G 5G; I mentioned it in my opening article. We also mentioned the Internet of Things value chain.
However, in 4G, we can say that ‘Mobile Video’ is the most basic and widespread value proposition of this technology, that only autonomous vehicles need 5G for 5G. So in a way it is not clear why and how to serve 5G to the market. Track your discourse assets on M2M, connect your machines to platforms and save money! The same rhetoric continues on the Internet of Things, which is desering me.
While all these controversies and uncertainties continue, the world has begun to invest significantly in a new technology (or ideology). Artificial Intelligence of Things (AIoT); English The Internet of Things with AI support or Artificial Intelligence of Things (Intelligent Objects).
What is Artificial Intelligence of Things (AIoT)?
Artificial Intelligence of Things (AIoT); In the Internet of Things, we can summarize the subject objects as equipping them with artificial intelligence, that is, it will come into our lives as ‘Artificial Intelligence of Things (Intelligent Objects)’.
If we ask what 5G will change, one of the first answers is: it will make AI widespread, it will be. In this sense, rapid progress in the field of artificial intelligence has reshaped many sectors, while the internet of things and edge unit mainstreaming have recently been at the top of trending technologies, at least globally.
Let’s look at a new technology called Artificial Intelligence of Things (AIoT) born in artificial intelligence and internet of things technologies.
What is AI of Objects (AIoT)?
AI of Things (AIoT) is a combination of artificial intelligence (AI) technologies with the Internet of Things (IoT) infrastructure to achieve more efficient Internet of Things applications, improve human-machine interactions, and improve data management and analytics.
It is the intersection of the Internet of Things (IoT) and Artificial Intelligence (AI). IoT is about connecting people, objects to platforms and other objects through communication. AI (AI) is based on collecting techniques/samples to improve and train the business intelligence of software to understand new data based on models of similar data. AIoT is a recent term that refers to the closet of IoT and AI systems. To generate useful data obtained by digitizing the physical world (IoT) and to gain insights from this data is artificial intelligence (AI), the meeting of these two objectives in a common denominator (AIoT).
What Does AI of Objects (AIoT) Do?
Machine learning, which is artificial intelligence (AI) technology thanks to its data management, analytics and decision-making capabilities, provides IoT applications/systems with the ability to learn from data. Converts IoT devices/edge sensors into “learning objects”. AI also transforms Internet of Things (IoT) data into useful information, making it easier to make better decisions.
Artificial intelligence of things (AIoT) is a basic technology that allows internet of things data to be provided as a service. (IoTDaaS:IoT Data as a Service, Internet of Things Data Offering Service)
AI of Objects (AIoT) Application Areas
Application areas can vary from smart home apps to autonomous vehicles. But right now, AI of Things application areas are very retail product-oriented. Many of these applications focus on cognitive analysm in consumer devices.
For example, according to information on the Mindcommerce site, Sharp is part of what they call “Smart Living”; In the name of “more responsive technology”, it mentions AIoT as “Human Oriented IoT = AIoT”.
IoT Analytics is in the most important IoT developments report for 2020; In January 2020, Chinese electronics manufacturer Xiaomi announced plans to invest at least $7.2 billion in 5G and artificial intelligence of things (AIOT) over the next 5 years. It is said that the new technology stream will direct individual and corporate IoT device investments such as smart TVs, drones, electric scooters, air purifiers, routers, security cameras, etc.
Companies like SAS, on the other hand; He notes that there are many possibilities beyond individual consumer products for AIoT-enabled corporate and industrial data, which they refer to in the slogans “From data collection to collective learning.” In the paragraph in Mindcommerce; Chief Technology Officer Oliver Schabenberger said that in a connected world, AI is used to automate tasks; IoT explains how data generated by end devices can be used.
AWS IoT; the following announcement is available on the website.
Superior AI integration: AWS brings AI and IoT together to make devices smarter. You can create models in the cloud and deploy them to devices where they run 2 times faster than other offers.
What happens if the Internet of Things and artificial intelligence technologies in Forbes merge? Article contains predictions about their usage areas.
The AI-powered camera system to be positioned in the store can recognize with facial recognition with the introduction of customers through the door, allowing them to make the right orientation. The system collects information subject to customer permission in areas such as genders, product preferences, traffic flow, etc., analyzes data to accurately predict consumer behavior, and then uses that information to make decisions about store operations, from marketing to product placement to other decisions. For example, if the system determines that the majority of customers entering the store are Generation Y, it can increase sales by offering product ads or in-store custom products that cater to that demographic. Smart cameras can identify shoppers and allow them to skip the payment stage, as in the Amazon Go store.
Drone Traffic Monitoring
In smart cities, there will be several practical use of AIoT, including traffic monitoring with drones. Congestion can be reduced if traffic can be monitored in real time and adjustments can be made to traffic flow. When drones are deployed to monitor a large area, they can transmit traffic data, and then AI can analyze data and make decisions on how best to alleviate traffic congestion with adjustments to the speed limits and timing of traffic lights without human intervention. ET City Brain, a product of Alibaba Cloud, optimizes the use of urban resources using AIoT. This system can detect accidents, illegal parks and change traffic lights to help ambulances reach patients who need help faster.
Another area where artificial intelligence and the internet of things intersect is smart office buildings. Some companies prefer to install a network of smart perimeter sensors in office buildings. These sensors can detect which personnel are where and adjust temperatures and lighting accordingly to improve energy efficiency. Another usage scenario is to control building access with intelligent building facial recognition technology. A combination of connected cameras and artificial intelligence that can compare real-time images with a database to determine who will be granted access to a building will be the AIoT app.
Fleet Management and Autonomous Vehicles
AIoT is used today in fleet management to help monitor the fleet’s vehicles, reduce fuel costs, monitor vehicle maintenance, and identify unsadned driver behavior. Through IoT devices such as GPS and other sensors, and an AI system, companies are better managing their fleets thanks to AIoT.
Another way AIoT is used today is in autonomous vehicles that use radars, sonar, GPS and cameras to collect data about Tesla’s driving conditions, and then an AI system to make decisions about the data collected by the devices’ internet.
Autonomous Delivery Robots
Similar to how AIoT is used with autonomous vehicles, autonomous delivery robots are an example of an AIoT application. Robots have sensors that gather information about the environment in which the robot passes and then make instant decisions about how to respond through the built-in AI platform.
Digitalization in Industry – Industry 4.0 AIIoT – Industrial AI of Things
At the heart of the examples mentioned above is centered around objects learning a task, observing environmental conditions and making decisions. Therefore, it will not be difficult to say that one of the most powerful uses will be Industry 4.0 applications.
With the substitution of sensors or machines that have completed their life in industrial conditions, we will witness examples that will allow the new device or machine to resume where the previous one left off.
AI of Objects (AIoT) Application Areas Summary
Artificial Intelligence of Things (AIoT); due to the intersection set with artificial intelligence (AI) and the Internet of Things (IoT), it will be placed in infrastructure components such as all connected programs, chipsets and edge unit analystics. ApIs here can then interoperability between components; used to provide at device level, software level and platform level. These units will focus primarily on optimizing system and network operations, as well as gaining value from data.
While the concept of AIoT is still relatively new, there are many possibilities for improving industry verticals such as the business, industrial and consumer products and service sectors. Real-time data is the core value of all AIoT usage scenarios and solutions. Real-time data also needs the capabilities of 5G.
Will AI of Things (AIoT) Be in the Cloud or at the Edge?
Before evaluating this question from both angles, let’s answer the question of why solution providers should integrate into IoT platforms.
There were M2M applications in our lives before IoT Platforms became peyda. These applications consist of cloud software developed specifically for verticals. A vertical, more importantly, for a usage scenario, you had to have a solution provider and the platform it provided. The more you expanded your usage scenario, the more likely you were to need more partners and cloud software.
What is being explained in this topology is the lack of interoperability and interdisciplinary data interaction in IoT-free IoT applications, which I call m2m applications.
Although its name is IoT in our country, it is the most preferred solution approach.
IoT Platform Applications
Applications and solutions that interact on a common IoT platform can feed each other and increase end-user benefit.
There are very few innovative solution providers in our country that have developed this model.
AI of Objects (AIoT) Must Be in Edge Host
Artificial intelligence support for internet of things technologies is among the scenarios in which it will be on the device side, that is, on the edge of the end unit in the field; autonomous vehicles and health detection/imaging usage scenarios will be at the fore. This can be preferred to prevent critical energy and communication latency in IoT Projects. For example, it can be fatal to have decision-making mechanisms on the end side so that an autonomous vehicle driving at 100 km/h can make its decisions faster.
AI of Objects (AIoT) Must Be on Cloud Platform
Bringing AI support to IoT platforms can strengthen the value prose of many IoT implementations. For example, site-positioned IoT projects for predictive maintenance and anomaly detection. Let’s have a project to monitor and monitor liquid tanks. It will be absurd that thousands of devices will work in different conditions in the field. Thousands of devices and environmental conditions in the field will allow you to learn the change of sensor responses and best up your software.
5G, IoT and AIoT Ecosystem Business Model
Here’s what I understand and try to tell you from this image: It is the development of a win-win strategy with complex business models intertwined with each other rather than the sale of boxes of new generation technologies.
The two regions where the arrows are focused actually point to telecom operators. That’s exactly what I tried to explain in our telecom operators and solution ecosystem article. Because telecom operators can be both a marketplace, i.e. a sales channel, and the IoT platform has the ability to offer software with high investment costs, such as Artificial Intelligence, to customers and business partners in the monthly model. While the IoT success of leading telecom operators globally has proven itself; You can find out why there are still deficiencies in our country in the details of our Article Digital Transformation and IoT Investments under covid-19 effect.
As a last word; Which IoT Communication Technology Should Be Selected? Let’s be redacted in this article as we have those that we have thed in our article. The next generation of technologies has dynamics beyond being a party or focusing on a single area and/or business model. For this reason, relevant investors should remember that these technologies are complementary when looking out for technology investments. The name of the business model to be included here will be competitive partnership. While companies in the same ecosystem are competitors, they add value to value propositions with the interaction of their solutions.