Volume 14 | Issue 6 |

 Abstract

The convergence of Artificial Intelligence (AI) and the Internet of Things (IoT) is transforming agriculture into a data-driven, automated, and sustainable domain. By integrating intelligent analytics with sensor-rich environments, smart agriculture aims to optimize crop yields, resource utilization, and food supply chains. This review critically examines recent developments in AI-enabled IoT for agriculture, highlighting architectures, algorithms, applications, and implementation challenges. We synthesize insights from more than forty-five scholarly works (2020-2025) and provide a thematic analysis of research trends. Particular attention is given to connectivity constraints, interoperability gaps, cybersecurity risks, and socio-economic barriers that hinder adoption. Comparative evaluations of existing IoT-AI frameworks demonstrate notable improvements in precision farming, yet reveal persistent limitations in scalability, explainability, and energy efficiency. The paper identifies research opportunities in edge/fog computing, federated learning, explainable AI (XAI), and blockchain-enabled trust frameworks. By offering a comprehensive synthesis of progress and open challenges, this review positions AI-driven IoT as a cornerstone for Agriculture 5.0 and provides guidance for researchers, policymakers, and practitioners working toward resilient, inclusive, and sustainable food systems.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Smart Agriculture, Artificial Intelligence, Internet of Things, Precision Farming, Machine Learning, Cybersecurity, Agriculture 5.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Urban traffic congestion significantly impacts emergency response times and public transport efficiency. Traditional fixed-time and semi-actuated signal systems fail to adapt dynamically to such priority needs, resulting in delayed emergency services and unreliable bus schedules. This project presents a Self-decisive traffic monitoring at crossroad junction system that integrates vehicle-in-urgency and public transport first preference mechanisms using IoT, sensors and vehicle-to-infrastructure (V2I) communication. The system detects approaching vehicle-in-urgency vehicles or buses via RFID, GPS, or camera-based sensing, predicts their arrival, and temporarily overrides normal signal cycles to grant a green corridor while ensuring pedestrian and cross-traffic safety. Conditional preference is also provided to delayed public transport vehicles to enhance schedule that follows. The system is assembled using Raspberry Pi 5 and the program code is executed using Thonny.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Smart Traffic Monitor and Control, Emergency Vehicle Pre-emption, Public Transport Priority, IoT and V2I Communication, Adaptive Signal Control, Intelligent Transportation System (ITS), Crossroad junction

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Water scarcity is becoming a serious issue in many urban areas due to increasing population and excessive consumption of freshwater resources. Grey water, which includes wastewater generated from household activities such as washing, bathing, and cleaning, can be reused for non-potable purposes if properly monitored. This paper proposes an IoT-based smart grey water management system using Raspberry Pi 4 to monitor water quality parameters such as pH and turbidity. The system automatically decides whether the water is reusable or not. If the water quality meets predefined standards, a Solenoid Valve directs the water to a reusable storage tank; otherwise, it is diverted to a waste tank. Sensor data is transmitted to the cloud platform ThingSpeak, allowing users to monitor water quality through a mobile device or laptop. The proposed system improves water conservation and enables intelligent water management in residential environments.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Grey Water, IoT, Water Quality Monitoring, Raspberry Pi, Turbidity Sensor.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Groundwater accumulation in mining areas leads to operational delays, increased maintenance costs, and safety hazards. Conventional dewatering systems rely on diesel-powered pumps or grid electricity, which are costly and environmentally harmful. This paper presents the design and implementation of a solar-powered automated dewatering system using an Arduino-based control mechanism. The system utilizes an ultrasonic sensor to monitor water levels and a relay-controlled DC pump for automatic water removal. A solar panel integrated with an MPPT charge controller provides a sustainable energy source, ensuring reliable operation in remote locations. To enhance system capabilities, an IoT-based monitoring approach is proposed, enabling real-time data visualization and remote access through cloud platforms such as ThingSpeak or Blynk. The proposed system aims to reduce human intervention, improve efficiency, and promote eco-friendly mining operations. Experimental observations indicate stable operation, efficient energy utilization, and effective automation, making the system suitable for small and medium-scale mining applications.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Solar Energy, Dewatering System, Mining Operations, IoT, Automation, Ultrasonic Sensor, Renewable Energy

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Electric fences are widely used in agricultural fields, industrial premises, and restricted areas for perimeter protection. However, unauthorized use, tampering, and accidental contact with electric fences can lead to safety hazards, property damage, and misuse of electrical energy. This paper presents an IoT-based smart hardware system designed to detect and prevent unauthorized use of electric fences through real-time monitoring, control, and alert mechanisms. The proposed system integrates a microcontroller-based unit with voltage and current sensors, motion detection modules, relay control circuits, and wireless communication technology such as Wi-Fi or GSM. The hardware continuously monitors fence status, power consumption, and intrusion attempts. If unauthorized activation, wire cutting, bypassing, or abnormal current flow is detected, the system immediately disconnects the fence power supply using an automated relay mechanism and sends instant alerts to the owner or security personnel through a mobile application or SMS.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

IOT, electric fence, unauthorized

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Autism Spectrum Disorder (ASD) is a neurodevelopmental condition characterized by impairments in social interaction, communication, and behavior, making early diagnosis crucial for effective intervention. Traditional diagnostic methods rely on behavioral assessments, which are often time-consuming and subjective. In recent years, deep learning-based computer vision techniques have emerged as promising tools for automated and non-invasive autism detection using facial images. This paper presents a comprehensive review of deep learning approaches for facial image-based autism detection, focusing on convolutional neural networks, transfer learning models, and hybrid architectures. The study analyzes commonly used datasets, evaluation metrics, and performance trends, highlighting that modern approaches achieve high accuracy but remain limited by dataset size, diversity, and clinical validation. Furthermore, key challenges such as ethical concerns, generalization issues, and lack of interpretability are discussed. The paper also outlines future research directions, including multimodal learning, explainable artificial intelligence, and the adoption of advanced architectures, to enhance the reliability and applicability of automated ASD detection systems

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Autism Spectrum Disorder, Deep Learning, Facial Image Analysis, Convolutional Neural Networks, Transfer Learning, Computer Vision, Medical Image Analysis, Explainable AI.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Memristor based systems represent an innovative approach to secure communications systems, nonlinear dynamics and hardware based efficient cryptographic systems. These systems demonstrate a unique combinatorial behavior produced by their non-linearity and ability to retain memory. As a result, they can exhibit a large variety of complex chaotic behaviors. When combined with field programmable gate arrays (FPGAs), memristor based chaotic systems may have the ability to be implemented in real-time, scalable and reconfigurable manner that makes them excellent candidates for current and future engineering applications. In this article, we review the literature pertaining to the generation of memristor based chaos on FPGA platforms through three major categories of discussion including theoretical foundations of memristors, implementation techniques for memristors and performance metrics of memristors from various study's performed within the literature. Performance metrics will be evaluated using comparison metrics including: hardware utilization, power consumption rate, throughput rate and system complexity. Finally, this research paper will identify the key gaps within existing research on memristor based chaos, outline limitations and discuss possibilities for future research. The combination of FPGA reconfigurability and memristor non-linear dynamics will create a new paradigm for the generation of high-speed, efficient and secure chaotic systems that can potentially be used in the future.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

FPGA, Memristor, Chaos, Cryptography, Throughput, Reconfigurability

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This project presents an IoT-based Smart Medication Management and Pill Dispenser System designed to promote medication adherence through automation, tracking, and remote supervision. The system ensures that patients receive the right dose at the right time, minimizing human errors and noncompliance. It integrates IoT technology, sensors, and cloud connectivity to automatically dispense medication, monitor intake, and alert both patients and caregivers. A companion mobile application provides real-time updates, reminders, and visualization. By bridging healthcare with smart automation, the system improves patient safety, enhances treatment outcomes, and supports the growing demand for digital healthcare management Introduction

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Internet of Things (IoT), Smart Medication System, Pill Dispenser, Medication Adherence, Automated Drug Delivery, Health Monitoring, Embedded Systems, ESP32 .

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Waste management has become a major global challenge due to rapid urbanization and increasing population. Improper segregation of waste at the source leads to environmental pollution, inefficient recycling, and health hazards. This paper proposes a Smart Waste Segregation and Recycling System using Internet of Things (IoT) technologies. The system utilizes multiple sensors such as moisture sensors, metal sensors, and IR proximity sensors to automatically classify waste into biodegradable, recyclable, and non-biodegradable categories. A motor-driven mechanism ensures proper segregation into designated compartments. Additionally, ultrasonic sensors monitor bin fill levels and send real-time alerts using IoT modules. The system also incorporates a data-driven reward mechanism to encourage public participation. The proposed solution enhances waste management efficiency, reduces manual labor, and promotes sustainable environmental practices.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

IoT, Waste Segregation, Smart Bin, Sensors, Recycling, Automation.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Proper medication intake is a major concern, particularly for elderly individuals and patients suffering from long-term illnesses. Missing doses or consuming incorrect medication can lead to severe health complications and increased medical risks. This paper proposes an Integrated Drug Dispenser system that automates the process of medicine distribution while improving patient safety and adherence.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Smart Medication System, Automated Drug Dispenser, Raspberry Pi, GSM Aler, Healthcare IoT, Elderly Assitance

  License

Creative Commons Attribution 4.0 and The Open Definition