Volume 14 | Issue 7 |

 Abstract

The growing demand for intelligent and automated solutions in daily life has led to the development of smart systems that enhance convenience, safety, and efficiency. One such critical area is infant care, where continuous monitoring and timely response are essential for ensuring the well-being of a baby. In many modern households, caregivers often face difficulty in maintaining constant supervision due to work commitments and other responsibilities. This situation creates a need for a reliable system that can assist in monitoring infants and responding appropriately without continuous human intervention. The proposed work presents the design and development of an AI-based smart cradle system that combines embedded technology, sensor networks, and intelligent decision-making to improve infant care.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Artificial Intelligence, Smart Cradle, Infant Monitoring, Internet of Things, Embedded Systems, Automation

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Road maintenance, particularly pothole repair, poses significant challenges for transportation authorities worldwide. Traditional methods of pothole detection and repair are often inefficient and costly. This project introduces a semi-automated robotic system for pothole identification and repair cost estimation, aiming to streamline road maintenance processes. The system employs an ultrasonic sensor for pothole detection, a GPS module for location tracking, and a Node MCU controller for data integration. An Arduino ATmega328 controller manages robot movement via servo and DC motors. The robot utilizes the depth measurements from the ultrasonic sensor to estimate repair costs based on the thickness of road layers. Real-time data is displayed on an LCD screen and transmitted wirelessly through Bluetooth and the Blynk IoT platform. Initial tests demonstrate the system's ability to accurately detect potholes, measure their depth, and estimate repair costs. The robot successfully navigates road surfaces, providing precise location data for each identified pothole. Cost estimations show a strong correlation with actual repair expenses, varying based on pothole depth and corresponding road layer thickness. This semi-automated approach offers a promising solution for enhancing road maintenance efficiency. By automating pothole detection and providing accurate cost estimates based on depth and road layer thickness, the system has the potential to optimize resource allocation and improve overall road quality. Future work will focus on expanding the system's capabilities and conducting large-scale field trials.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Pothole detection, Semi-automated robot, IoT, Ultrasonic sensing, GPS mapping, Repair cost estimation, Road maintenance, Infrastructure management.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The EV Battery Management System using Arduino Uno is designed to monitor and protect battery performance. It tracks voltage, current, and temperature to prevent overcharging, deep discharge, and overheating. The system also includes fire protection through automatic cutoff during abnormal conditions. This ensures improved safety, efficiency, and longer battery life for electric vehicle applications. The EV Battery Management System using Arduino Uno monitors voltage, current, and temperature to ensure safe operation. It prevents overcharging, deep discharge, and overheating, and includes fire protection with automatic cutoff. The system improves battery safety and lifespan.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Arduino, Bomb Disposal Robot, Flex Sensor, Human Motion Synchronization, RF Communication, Robotic Arm, Servo Motor, Wireless Control,Artificial Intelligence

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Flood-related vehicle accidents increase every year during heavy rainfall and urban waterlogging. Many drivers fail to identify dangerous water depth while driving, which leads to vehicle damage, passenger trapping, and loss of life. This project presents a smart car safety system designed for emergency flood situations using water level sensors, ultrasonic sensors, GPS, GSM, and a microcontroller-based control unit. The system continuously monitors water levels around the vehicle and alerts the driver when the water reaches unsafe conditions. If the water level crosses a critical threshold, the system activates automatic safety measures such as engine cutoff warning, emergency alerts, location tracking, and rescue notifications.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Flood Detection, Vehicle Safety, IoT, Water Level Sensor, GSM Module, GPS Tracking, Emergency Alert System, Smart Transportation

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The trend of increasing globalization in the distribu-tion of pharmaceuticals has brought a pressure to secure authen-tication and effective tracking of the supply chain to ensure the safety of drugs and compliance with regulations. Traditional ways of monitoring often have low levels of transparency, counterfeit is detected late, management of data is spread over, and there is poor inventory forecasting. The proposed architecture utilizes the data provided by numerous points in the pharmaceutical supply chain, such as the pharmaceuticals stock details from the user, blockchain communication records, supply and inventory management systems, and the records of geolocation tracking collected during the operations conducted at the global scale. Ad-vanced missing value imputation, multivariate outlier detection, PCA as well as feature priority ranking are all part of fuzzy data preprocessing to enhance data quality and reduce dimensionality. RF, SVM, are among the ML models that are combined with hybrid ensemble strategies that include as stacking, blending and weighted voting that are linked with Ethereum-based smart contract. The metrics that are used to measure model perfor-mance are accuracy, precision, recall, F1-score and ROC-AUC. The combined ensemble architecture comprising of the Random Forest performed better with an accuracy of 96.1as that inventory optimization lowered carrying costs by 23 percent and inventory stockout cases by 67 percent. The unified architecture enhances significantly real-time authentication, supply chain visibility and scalability of operations.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Demand Forecasting, Anomaly Detection, Phar-maceutical Inventory Management, Role-Based Access Control, Inventory Monitoring, Healthcare Systems

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The rapid growth of wireless communication and Internet of Things (IoT) devices has increased the need for secure data transmission, especially in resource-constrained embedded systems. Traditional encryption algorithms such as AES and DES provide strong security but require high computational power and memory, making them unsuitable for 8-bit microcontrollers. This project presents a lightweight encryption and decryption system based on an 8-bit RISC processor for secure communication. The proposed system uses a hybrid encryption technique that combines Affine transformation, bit rotation, and key feedback mechanisms to ensure data confidentiality while maintaining low computational complexity. The ESP8266 Wi-Fi module is used to receive user input through a web interface and perform encryption. The encrypted data is then transmitted to an Arduino UNO, which acts as an 8-bit RISC processor and performs decryption using inverse operations. The system enables real time communication through serial (UART) transmission and displays both encrypted and decrypted data on an LCD module. The implementation demonstrates that secure communication can be achieved efficiently using simple mathematical operations on low cost and low power embedded platforms. This project highlights the feasibility of lightweight cryptographic techniques for applications in IoT and embedded systems where resources are limited. Keywords: 8-bit RISC processor, Encryption and Decryption, Lightweight Cryptography, Affine Transformation, Bit Rotation, Key Feedback Mechanism, ESP8266 Wi-Fi module, Arduino UNO, Secure Data Transmission, Embedded Systems, Internet of Things (IoT), and UART communication.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Encryption and Decryption, Lightweight Cryptography, Affine Transformation, Bit Rotation, Key Feedback Mechanism, ESP8266 Wi-Fi module, Secure Data Transmission, Internet of Things (IoT), and UART communication.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The rapid growth of e-commerce platforms has generated massive amounts of customer behavior data, offering opportunities to predict online purchasing decisions. Understanding customer behavior not only supports personalization but also enhances customer retention and increases revenue. This study applies machine learning techniques to analyze and predict customer purchase behavior in online retail settings. Models such as Logistic Regression, Decision Tree, Random Forest, Support Vector Machine (SVM), and K- Nearest Neighbors (KNN) were implemented and evaluated. The proposed methodology leverages feature engineering, preprocessing, and model evaluation using accuracy, precision, recall, F1-score, and ROC-AUC curves. Experimental results demonstrate that ensemble methods like Random Forest provide superior predictive performance, achieving an accuracy of 89% with an ROC-AUC of 0.92.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Machine learning, E-commerce, Purchase Trends, Personalized recommendations

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Women's safety has become a major concern due to the increasing number of harassment and emergency incidents occurring in both public and private spaces. Conventional safety systems often suffer from limitations such as delayed communication, dependence on cellular networks, and unreliable location tracking in indoor environments. To address these challenges, this paper presents the design and implementation of the Promise Pendant, a compact wearable safety device developed for rapid emergency response and real-time communication. The proposed system integrates an Arduino Uno, ESP32-CAM module, panic button, voice-assisted activation, and Telegram-based alert mechanism. During emergency situations, the device sends instant alert messages with approximate location details and captured images to predefined contacts through internet connectivity. Initially, GPS and GSM modules were considered; however, due to poor indoor performance, increased hardware complexity, and higher power consumption, the system was redesigned using IP-based location identification and Telegram communication.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

AI wearable, Emotional intelligence, GPS tracking, Smart jewelry.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

This research presents a comprehensive and detailed exploration into the design, structural development, and hardware implementation of a hybrid electric bicycle featuring an innovative multistage regenerative braking system (RBS). In the contemporary era of rapid urbanization and escalating environmental challenges, the fundamental objective of this study is to significantly optimize the energy efficiency and operational sustainability of lightweight electric mobility solutions. This is achieved by effectively harvesting and repurposing kinetic energy that is traditionally dissipated and lost as waste heat during periods of deceleration, stop-and-go city transit, and downhill travel. While conventional electric bicycles and small-scale electric vehicles rely almost exclusively on mechanical friction brakes - a process that represents a substantial loss of potential power - the proposed architecture replaces this inefficient method with a precision-controlled electromagnetic braking mechanism governed by an Arduino Nano microcontroller.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

DESIGN AND DEVELOPMENT OF HYBRID E-BICYCLE WITH MULTISTAGE REGENERATIVE BRAKING

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Dealing with kitchen waste is one of the biggest headaches in urban living. Usually, it just ends up in a landfill, which is a massive waste of nutrients that could be put back into the earth. To fix this, we've developed a compact, automated system designed specifically for homes and restaurants that turns food scraps into plant food instantly. Most composting takes forever and requires a lot of manual labor. Our system skips the wait by using a dual-motor setup to handle everything: The Grinder: A high-torque motor shreds various types of wet waste into smooth, consistent slurry. The Distributor: A second motor pumps that nutrient-rich liquid directly into garden soil or planters.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Automated Waste Management, Wet Waste Crusher, Nutrient Recovery, Urban Farming, Bio-slurry, and Decentralized Recycling

  License

Creative Commons Attribution 4.0 and The Open Definition