Research Trends in Artificial Intelligence: Internet of Things

The “Smart Farm” project is an IoT-based agriculture project aimed at optimizing crop yield and promoting sustainability in farming practices. By integrating various IoT devices and sensors, the project aims to improve the efficiency of farming operations, reduce waste, and enhance the quality and quantity of crop yields. The project focuses on using IoT technology to monitor and control various aspects of farming, including soil moisture, temperature, humidity, crop health, and livestock behaviour. By leveraging data from these sensors and devices, farmers can make more informed decisions about irrigation, pest control, and crop management, leading to increased productivity and sustainability. Overall, the Smart Farm project seeks to transform traditional farming practices into a more efficient, data-driven, and sustainable model that benefits both farmers and the environment.

Every nation is concerned about the growing problem of agriculture automation. It is challenging to supply the food needs of the existing population due to rising numbers, frequent climate change, and scarce resources. Farmers are forced to wreak havoc on the land by applying dangerous pesticides more often since their old techniques cannot keep up with the growing demand. As a result, agricultural practices are significantly impacted, and the land gradually loses its fertility and becomes unproductive. The agriculture sector can benefit from technology like Artificial Intelligence (AI), deep learning, the Internet of Things (IoT), embedded systems, and automation. Artificial neural networks, the Internet of Things, fuzzy logic, machine learning, and other technologies may all be used to automate agricultural systems. Artificial intelligence technology is advancing quickly, and as a result, its employment is in a wide range of fields. Utilizing clever technologies, the agricultural industry has become able to regulate the field environment that is essential to the care of every plant. A suitable atmosphere and appropriate irrigation are provided by the plant's identification and suitable circumstances. In order to increase agriculture yields, it has become important to manage crops in controlled settings like greenhouses that can enhance the output. This chapter focuses on the use of artificial intelligence and IoT technology to improve the productivity of agricultural enterprises. AI technologies might help farmers overcome problems like weeds, pests, and climatic variability that lower output. Numerous uses of AI are now being deployed, such as automatic machine changes for weather forecasting and pest detection. The goal of implementing AI and IoT is to increase the possibility of producing healthy crops by recognizing damaged crops and crop yield growth.

Technology such as the Internet of Things (IoT), big data, cloud computing, fog computing, edge computing, and blockchain can be a perilous factor when it comes to encouraging the integration of new technologies. Artificial Intelligence (AI) is an integral part of agricultural and industrial development. This chapter describes the extensive evolution of AI and IoT. Researchers and practitioners will find this chapter essential for understanding AI and IoT, along with models, current status, future trends, and industrial development. There are a number of issues with AI, but overall, it is considered to be an advanced and revolutionary assistant in a wide range of fields. The purpose of this chapter is to present a comprehensive study on AIoT that explains the convergence of AI and IoT. Herein, we summarize some innovative AIoT applications that are likely too intense for our world. 

India's GDP is heavily reliant on agricultural products and business management. Therefore, it is crucial for the agriculture industry to comprehend the most common uses of artificial intelligence (AI) through case studies. To increase its production, this industry must overcome a number of obstacles, such as soil treatment, plant disease and pest effects, crop management, farmers' innovative methods, and the use of technology. The major ideas behind AI in agriculture are its adaptability, excellence, accuracy, and economy. It is critical to examine AI applications for managing soil, crops, and the environment, and plant or leaf diseases. Food security continues to be seriously threatened by deforestation and poor soil conditions, both of which harm the economy. The application's advantages, constraints, and methods for employing expert systems to increase productivity are all given particular attention. Businesses are utilizing robots and automation to assist farmers in developing more effective weed control strategies for their crops. See & Spray, a robot created by Blue River Technology, is said to use computer vision to monitor and accurately spray weeds on cotton plants. Crop and Soil Monitoring - Businesses are using deep learning and computer vision algorithms to interpret data taken by drones and/or software-based technologies to monitor the health of crops and soil. Crop sustainability and weather forecasting are accomplished via satellite systems. A Colorado-based startup employs satellites and machine learning algorithms to examine agricultural sustainability, forecast weather, and assess farms for the presence of diseases and pests. Utilizing predictive analytics, machine learning models are being created to monitor and forecast various environmental factors, such as weather variations. Drones and computer vision are used for crop analysis, while machine learning is used for identifying soil flaws.

Diabetes prediction is an ongoing research problem. The sooner diabetes is detected in a human, the sooner lives and medical resources can be saved. Predicting diabetes as early as possible with easy to measures parameters with optimal accuracy is an ongoing problem. When dealing with large data, feature selection plays an important role. It not only reduces the computational cost but also increases the performance of a model. This study ensemble three different types of feature selection techniques: filter, wrapper and embedded. Ensembling Boruta and LASSO features give optimal results. Also, effectively handling class imbalance leads to better results.

Everyone has been utilizing the Internet of Things in current years (IoT). Therefore, as the IoT has grown enormously, so too have concerns about Artificial Intelligence. Artificial intelligence (AI) is at the forefront of ubiquitous computing and is utilized to create intricate algorithms to safeguard networks and systems, including IoT devices. But to carry out cyber security assaults, hackers have learned how to make use of AI, and they have even started to deploy antagonistic AI. The goal of this research work is to comprehensively present and summarize the pertinent literature in these fields. It does this by compiling data from a few other surveys and research papers regarding sustainable development using AI, IoT, and attacks with and against AI. It also shows the relationship between IoT and AI.

Digital twin technology is an important part of the industry 4.0 revolution. Digital twins is a concept of integrating different smart technologies including integration of data and digitization. The vision of the digital twin technology is based on the philosophy that any component, assembly, system, process, product, or even environment can be replicated in terms of form, functionality and several other parameters in a digital way throughout different phases of its lifecycle. The Digital Twin concept is based on the highly growing information and communication technologies alongside conventional methods for getting better interconnection and integration amongst all the entities involved in a particular phase of the product lifecycle. The Internet of things (IOT) and artificial intelligence (AI) are crucial parts of the Industry 4.0 revolution. IOT proposes to embed electronics, sensors, network connectivity and different software platforms with products. Better integration between the physical and digital world is achieved by a strong network infrastructure that facilitates remote sensing, monitoring and even controlling of connected systems. This technology provides users with many benefits such as increased efficiency, reduced downtime, improvement in precision and accuracy along with cost saving. This chapter includes an overview of digital twin-enabling technologies and their applications. The applications range from Artificial Intelligence, Internet of Things (IoT) to manufacturing.

The volume of data is increasing quickly in the modern day. Effective information retrieval techniques are needed to extract important facts from such a large collection of information. As a result, retrieval of information is the process of gathering valid data from a variety of sources. The majority of the time, information is retrieved from the internet using search queries. The aim of this research is to explore various issues existing in information retrieval techniques and to propose new techniques to overcome existing challenges in the field of Information retrieval. Modern information retrieval methods have been examined, and it was discovered that they do not take semantic keyword knowledge into account when returning results. The semantic web is a development of the internet that enables computers to comprehend human inquiries in terms of their intent and produce pertinent responses. This research mainly focuses on Ontology-Based Information Retrieval which can support semantic similarity and retain the view of an approximate search in a document repository using machine learning techniques. Further, this research works explores an adaptive update model for retrieving the information and proposes a semantic search model for the given user query. The objective of ontology-based semantic web information search is to increase the accuracy, precision and recall of user queries.

The COVID-19 epidemic has completely altered the environment and every aspect of every individual. The most affected part is the education system and the stakeholders associated with it. Organizations are currently being forced to adapt and alter their strategies in response to the new situation created by the COVID-19 epidemic. The proposed study gathers tweets on online schooling from social media sites like Twitter and Facebook comments in order to conduct a thorough sentiment analysis (SA) during the epidemic. The current study utilizes techniques for natural language processing (NLP) and machine learning (ML) to extract subjective data, establish polarity, and identify how people felt about the educational system prior to and following the COVID-19 crisis. The first step in the proposed study is to retrieve tweets using Twitter APIs before they are ready for rigorous preprocessing. One filtering method is Information Gain (IG). We will identify and examine the latent causes of the unpleasant feelings. We'll look at the machine-learning classification algorithm at the end. The proposed model will analyse the perceptions of people about the online educational system during COVID-19

The automation industry is rapidly growing and coming up with new and improved techniques for reducing time and efforts. One such example is the autonomous cars which are said to be the future of the automobile industry since they would be driver less, very efficient and relieve the stress of daily commuting [1]. Advances in technology using the AI and deep learning techniques help in improving the safety of the passengers and also in minimizing the efforts of the driver. For the study of autonomous vehicles, a lot of data needs to be collected, some of which include warning signals, speed limits, obstacles, collision avoidance, etc. This paper shows how IoT devices i.e. cameras and LiDAR sensors help in data collection, how deep learning is a solution, and how image recognition methods that use deep learning can help in object or any obstacle detection. An image processing algorithm based on deep learning is proposed in which the image perception can be made by an optical camera communication technique that can be used for collecting the data. Hence it will highlight how deep learning is used in the field of image processing or image recognition.

Due to the rapid increase in urban population, the today’s life of every citizen undergoes drastic changes. For the betterment of human life, Government of India had decided and announced the development of smart cities i.e the cities to be developed with all modern facilities in which people can use the internet for their daily activities. Smart city development would heavily depend on the Internet of Things (IOT) which combines three important aspects of Internet such as people-people interaction, people-objects interaction, and objects-objects interaction. Artificial Intelligence (AI) is another technology that provides the city equipped with efficient systems for security, safety, parking, etc. The applications of AI enable IoT in smart cities that are discussed in this chapter such as Smart Home, Smart Healthcare, Smart Water, Smart Grid and Energy, Smart Transport, and Real Estate investment. Section 12.1 gives an introduction about the chapter that includes an introduction to Smart City, IoT and integration of AI and IoT. Section 12.2 discusses the potential use cases of AI and IoT in smart cities. Smart Home use case that includes smart management of equipment and human activity recognition is discussed. It also discusses Smart Healthcare which includes the fitness-tracking system, glucose-level monitoring system, body-temperature monitoring system, stress detection system, and oxygen-saturation monitoring system. Smart Infrastructure is also discussed which includes Smart Water, Smart Grid and Energy, and structural health monitoring. Smart Transport is also discussed in this chapter which includes the Vehicle Infrastructure Pedestrian (VIP), Smart Parking System (SPS) and Automated Incident Detection system (AID). The final section concludes the chapter by discussing the challenges in the smart city environment and future enhancements.

This paper presents a Virtual Trial Room software using Augmented Reality which allows the user to wear clothes virtually by superimposing 3d clothes over the user. These sensors are valued particularly for robotics or computer vision applications because of their low cost and their ability to measure distances at a high frame rate. In November 2010, the Kinect v1 (Microsoft) release encouraged the use of Red Green Blue (RGB)-D cameras, and in July 2014, a second version of the sensor was launched. Because high-frequency point nuclei can be obtained from an observed picture, users can imagine employing these sensors to fulfill 3D acquisition requirements. However, certain issues such as the adequacy and accuracy of RGB-D cameras in close-range 3D modeling have to be addressed owing to the technology involved. The quality of the data obtained therefore constitutes an important dimension. In this study, the usage of the current sensor Kinect v2 is explored in the three-dimensional reconstruction of tiny objects. The advantages and problems of Kinect v2 are addressed in the first section and then photogrammetry versions are presented after an accurate evaluation of the generated models.