• Home
  • 2019
  • September

Mingling of IoT with AI

All we need to do to create an accurate and working Machine Learning models is a huge amount of high quality and relevant data. Internet of Things (IoT) devices have the potential to generate a vast amount of data which can be then used with AI. Imagine facial recognition systems using cameras to replace ordinary payments, or recall the current hype for autonomous cars gathering data about surroundings using built-in sensors. Those solutions use both AI and IoT, yet little has been written on how to easily integrate them together. Fortunately, existing IoT platforms provides the interface to gather the data from various devices and can offer a relatively easy way to utilize the IoT data into AI/ML systems. This article will provide you with a few state-of-the-art AI+IoT examples, an overview of the most popular IoT platforms and how they can be integrated with your AI/ML systems.

Just to give you a better feel of how AI can be utilized with IoT, here present two business cases which use both technologies, and a list of some of the most popular IoT platforms with AI capabilities. 

ET City Brain

ET City Brain, created by Alibaba Cloud, is a complex AI platform solution which optimizes usage of urban public resources. It has been successfully implemented in Hangzhou, China leading to a decrease in car traffics by 15%. The solution also helps to detect road accidents, illegal parking and supports ambulances to reach their target faster by changing the traffic lights to help it reach the patient faster. It is an outstanding software system which utilizes traffic light cameras from the whole city and based on the output of machine learning models it can determine how traffic lights should be changed.

Tesla's Autopilot

Source: Tesla

Tesla autopilot system incorporates GPS, cameras, sonars and forward-looking radars together with specialized hardware to fully utilize the usage of the data coupled into Neural Networks architecture. It works as a self-enclosed system which gathers the data from sensors and then uses a Neural Network model to help determine what should be the next change in the movement of the car .

IoT Platforms with AI capabilities

IoT platforms are getting increasingly popular as they offer a variety of useful tools and cloud storage options for the data streamed from IoT devices. What does each platform offer?

Microsoft Azure IoT Platform

Azure IoT has a Github repository which was created to help you integrate AI and IoT easily. This code repository provides different code examples on how to use Machine Learning and Azure IoT Edge together. Deep Learning models can be packaged in Azure IoT Edge-compatible Docker containers and then exposed to the data using REST APIs. Alternatively, it can be deployed on the cloud and then used for predictions with REST APIs. Azure IoT has a lot of useful tutorials and case examples which are available on their website.

Google Cloud IoT

Google Cloud IoT is an impressive IoT platform which enables you to connect the data to the machine learning model in many different ways. It offers a complete set of tools for edge/on-premises computing with machine learning capabilities, similarly to Azure IoT. Google has also created a separate AI Platform so you can train the model there, together with a cloud storage option for the IoT data. There are few useful tutorials out there: this one, for example, explains how to deploy the ML model to the IoT device which updates itself on the new upcoming data.


The well-known AWS also offers AI+IoT solutions. AWS IoT Analytics is especially interesting, as it offers data preparation, visualization, and machine learning tools. Machine learning model can be trained using Amazon Sagemaker and then containerized so it can be used with the data stream from IoT devices. ML models can also be uploaded to devices directly where they run much faster and they can also be automatically updated based on the new incoming data.

In conclusion, IoT is making machine connected and communicates with the entire network. IoT also generates Big Data, but AI is only the technology that makes those Big Data useful and meaningful for an industry. A reciprocally beneficial coexistence occurs between IoT and AI technologies. There are tons of domains and business niches, which can reap the advantages of the coexistence of both technologies.

Source: towardsdatascience


What is the Fourth Industrial Revolution?

Described as a range of new technologies that are fusing the physical, digital and biological worlds, the Fourth Industrial Revolution (4IR) is certain to alter the way the human race lives, works, and relates to one another.

A number of technological fields will see major advances over the next few years that will affect all disciplines, economies and industries. These fields include robotics, artificial intelligence, nanotechnology, quantum computing, biotechnology, the Internet of Things, 3D printing, autonomous vehicles, and more.

Experts predict a number of changes to come with 4IR. Over 7 million jobs will be affected over the next five years in the world’s largest economies, as technological progress in 3D printing and robotics starts to disrupt manufacturing and other industries.

Preparing for 4IR Begins with Education

Education is key for adapting to the changes 4IR technologies will bring. The employment landscape will undergo a massive shift, making advanced skills increasingly important. IEEE is helping to educate future leaders on their role in fostering innovation and shaping technological breakthroughs.

As the importance of understanding 4IR grows, IEEE publications have become a valuable resource for students and teachers to learn about emerging technologies. Every year IEEE publishes new journals, conference papers and standards with knowledge and insights that are helping to shape this revolution. A recent evaluation of the top ranked engineering and technology universities worldwide found that all top 100 schools subscribe to IEEE content and depend on IEEE information to educate future innovators (Times Higher Education).

To read the full article click here and to see the IEEE publications across all 4IR technologies, visit the IEEE Xplore digital library.


Top IoT Development Tools

The Internet of Things is penetrating every aspect of our daily life.  With this humongous interest, numerous organizations are battling a war against one another and are attempting to create products that are superior to their partners. A wave of IoT Tools has emerged over the past years. Here a list of the most Popular IoT development Tools are listed. 


Arduino is an open-source prototyping and simple-to-use IoT platform. The Arduino  uses microcontrollers programmed using any of the supported languages, C and C++.



This IoT IDE is created for Raspberry Pi board by IoT tech enthusiasts. With over 35,000 packages and numerous examples along rapid installation with the use of pre-compiled software makes it an essential IoT development tools.


Node-RED is a simple visual tool which is built on Node.js, a server-side JavaScript platform which is widely used in IoT projects. It is an open-source tool mainly used to connect devices, services and APIs together with an integrated browser-based flow editor.


This is a hardware provider that can be used to build basic IoT solutions and prototypes. Tessel 2 lends a helping hand through its numerous sensors and modules. This is a board which can hold up to a dozen modules including the RFID, camera, GPS and the accelerometer.


 This platform comes with a build system, supported by a library manager and IDE. It comes pre-equipped with more than 10 frameworks, more than 20 development platforms, and more than 400 embedded boards. It has support for C and C++ Intelligent Code Completion and Smart Code Linter for Professional Development. It also has support for multiple projects workflow in multiple panes.


Read more information here and here.

Open Standards for the Internet of Things

The Internet of Things, commonly abbreviated as IoT is a novel paradigm that describes the idea of everyday physical objects being connected to the Internet. This includes everything from mobiles, vehicles, air conditioners, headphones, wearable devices and almost anything. Unquestionably, the IoT idea has a high impact on several aspects of everyday life and behavior of potential users. Consequently, companies have started to introduce numerous IoT based products and services and several consortiums have been formed to define protocols and standards for the IoT. 

The article  “Open Standards for the Internet of Things” aims to provide a brief revision of the state of the art in the IoT field (5G IoT Standards, FIWARE), and a summary of some of the research proyects related to the IoT concept in which the Audio and Communications Signal Processing Group and the Multimedia Communications Group of the ITEAM UPV are currently involved (MAtchUP project and SSEnCe Project on Acoustic-aided IoT) . 

MAtchUP is an EU-funded Smart City project involving three lighthouse cities (Valencia-Spain, Dresden-Germany and Antalya-Turkey) and four follower cities (Ostend-Belgium, Herzliya-Israel, SkopjeFYROM and Kerava-Finland). It started on September 2018 and will finish on September 2023. During this period, MAtchUP partners will create and adopt solutions that can turn urban problems into smart opportunities to improve the citizens’ quality of life and boost the local economies, mainly in the areas of energy, mobility and ICT (information and Communications Technologies). The final aim is to create a prosperous and more liveable urban environment for communities.

The SSEnCe project (Sound-Aided Smart Environments for the City, Home and Nature) aims to encourage the dissemination and develop of real and practical prototypes focused on the Global concept of Intelligence in the IoT, particularly the applications are based mainly on the acoustic information of the environment. The main objectives of this project are the creation of an Observatory and the development of three technological demonstrators of immediate practical application.

Read the full publication here.