DEPARTMENT OF MECHATRONICS ENGINEERING

This paper critically examines the emerging field of blind computation in AI machine operations, challenging conventional approaches to data privacy and security in artificial intelligence systems. As AI continues to permeate various sectors, from healthcare to finance, the need for robust privacy-preserving techniques has become paramount. Blind computation offers a promising solution by enabling AI models to process encrypted data without decryption, thus maintaining data confidentiality throughout the computational pipeline. This research synthesizes cutting-edge developments in homomorphic encryption, secure multiparty computation, and federated learning, presenting a comprehensive framework for implementing blind computation in AI systems. We propose novel architectures that significantly enhance data protection without compromising computational efficiency. Our analysis reveals that while blind computation techniques offer unprecedented levels of privacy, they also introduce new challenges in terms of computational overhead and model accuracy. We present empirical evidence demonstrating the trade-offs between privacy, performance, and precision, and propose innovative strategies to optimize these competing factors. Furthermore, we critically assess the ethical implications of blind computation, examining its potential to either mitigate or exacerbate existing biases in AI systems. This paper concludes by outlining a roadmap for future research, emphasizing the need for interdisciplinary collaboration to address the technical, ethical, and regulatory challenges associated with blind computation in AI. Our findings have significant implications for the design and deployment of privacy-preserving AI systems across various domains, potentially revolutionizing the way sensitive data is processed in the age of artificial intelligence

This project presents the developments of a vision-based system for aerial pipeline Rightof-Way surveillance using optical/Infrared sensors mounted on Unmanned Aerial Vehicles (UAV). The aim of research is to develop a highly automated, on-board system for detecting and following the pipelines; while simultaneously detecting any third-party interference. The proposed approach of using a UAV platform could potentially reduce the cost of monitoring and surveying pipelines when compared to manned aircraft. To evaluate the performance of the system, the algorithms were coded using Python programming language. The pipeline endpoints are identified by transforming the 16-bits depth data of the explored environment into 3D point clouds world coordinates. Then, using the Random Sample Consensus (RANSAC) approach, the foreground and background are separated based on the transformed 3D point cloud to extract the plane that corresponds to the ground. Following that, these boundaries were filtered out, after being transformed into a 3D point cloud, based on the real height of the pipeline for fast and accurate measurements using a Euclidean distance of each boundary point, relative to the plane of the ground extracted previously. The filtered boundaries were used to detect the straight lines of the object boundary (Hough lines), once transformed into 16-bit depth data, using a Hough transform method. The pipeline is verified by estimating a centre line segment, using a 3D point cloud of each pair of the Hough line segments, (transformed into 3D). Then, the corresponding linearity of the pipeline points cloud is filtered within the width of the pipeline using Euclidean distance in the foreground point cloud. Then, the segment length of the detected centre line is enhanced to match the exact pipeline segment by extending it along the filtered point cloud of the pipeline. The third-party interference is detected based on four parameters, namely: foreground depth

This project presents the study and design of a full-scale gesture-controlled lift system that integrates advanced hardware and intelligent control strategies to achieve touch-free vertical transportation. Leveraging a PAJ7620 gesture sensor and an ESP32 microcontroller, the system interprets user hand gestures to control a robust motor/pulley mechanism designed to handle loads of up to 500 kg over a lift height of 10 meters. The design incorporates rigorous load calculations, safety features including emergency braking and limit switches, and a comprehensive structural framework built from high-strength steel and aluminum. Through extensive prototype development, testing, and optimization, the system demonstrated high reliability, efficient energy use, and ease of integration within existing lift infrastructures. This innovative approach not only enhances user accessibility and safety but also paves the way for future advancements in
automated, gesture-based control systems in commercial and industrial applications.

Pipeline infrastructure plays a critical role in Nigeria‘s energy sector, yet persistent challenges like unauthorized third-party activities threaten its security, reliability, and efficiency. This project focuses on improving pipeline monitoring systems, particularly in the context of Nigeria‘s challenges with third-party interference like vandalism and oil bunkering. The research examines existing monitoring methods (fiber-optic sensing, acoustic leak detection, real-time surveillance, and SCADA), analyzes successful global pipeline monitoring strategies (Trans Alaska and Nord Stream) and addresses the limitations of current monitoring systems. The study proposes an Improved Hybrid Pipeline Protection System (IHPPS) that integrates multiple technologies (flow measurement sensors, acoustic monitoring, GPS tracking, and IoT-based remote surveillance). The IHPPS aims to enhance detection accuracy, minimize false alarms, and improve real-time response, ultimately reducing economic, environmental, and security risks. The study concludes that a multi-layered monitoring system is crucial for improving pipeline reliability and security.

Ineffective waste management in rapidly urbanizing cities has intensified environmental and public health challenges, particularly in Nigeria, where traditional collection methods remain inefficient and costly. The emergence of smart waste bin technologies offers a sustainable alternative by integrating Internet of Things (IoT) components for real-time monitoring, automated waste handling, and optimized collection processes. This study addresses the limitations of existing systems—such as lack of automation, poor material sustainability, and insufficient adaptability to environmental conditions—by developing a smart waste bin prototype that enhances efficiency, reduces health risks, and supports global sustainability goals through innovation in sensor integration, automation, and eco-friendly design. This study encompasses the design and development of a smart waste bin system integrating sensors, automation, and wireless communication technologies. The prototype includes real-time monitoring, SMS notifications, and automated features such as waste compaction and lid control to enhance hygiene and usability. Emphasis is also placed on the use of sustainable construction materials and the system’s adaptability to various environmental conditions. Rigorous hardware and software testing ensures reliable sensor performance, effective automation, and accurate communication. These evaluations validate the system’s functionality in real-world scenarios, reinforcing its potential as a scalable and sustainable solution for modern waste management challenges. The developed smart waste bin system demonstrated successful integration of real-time monitoring, automation, and wireless communication technologies. Testing confirmed the reliability of its components—ultrasonic and load sensors, linear actuators, and GSM/GPS modules—under realistic operating conditions. The system effectively detected fill levels, triggered automated compaction, and sent timely alerts, thereby reducing overflow incidents and improving waste collection efficiency. These results suggest the system’s potential to address key urban waste management challenges, offering a scalable and sustainable solution adaptable to residential, commercial, and municipal applications

Vehicle theft and unauthorized access to people’s vehicles remain critical challenges worldwide, with developing countries such as Nigeria experiencing a rising incidence of automobile-related crimes due to inadequate security measures and poor enforcement of tracking technologies. Conventional anti-theft solutions such as alarms, mechanical locks, and immobilizers often fail against organized theft operations using signal jammers and key cloning devices. This project focuses on the simulated design of a real-time vehicle tracking and remote engine shutdown system using MATLAB Simulink and proteus as cost-effective development platforms. The proposed system integrates three primary components: a GPS module for continuous vehicle location monitoring, a GSM communication link for transmitting control signals, and a relay-based engine cutoff mechanism to remotely immobilize the vehicle when unauthorized movement is detected. The simulation model evaluates the system’s performance in terms of location accuracy, communication efficiency, and response speed, which are critical factors in environments where GSM coverage can be inconsistent, as is common in several regions of Nigeria. By leveraging Simulink’s block-based modeling, the design eliminates the need for immediate physical prototyping, reducing costs while allowing early validation of functional behavior. Results indicate that the system can track vehicle position accurately and execute engine shutdown commands within seconds, offering a practical solution for private vehicle owners, logistics companies, and government agencies managing various transportation fleets. This study demonstrates the potential of integrating real-time tracking with active control to enhance vehicle security in Nigeria and similar developing markets. It further provides a scalable platform for future extensions, including Internet of Things (IoT) connectivity, cloud data storage, and encrypted communication protocols to counter network vulnerabilities and improve reliability under real-world condition.

Automated gate systems have become essential in modern residential security due to the need for controlled access and reduced manual operation. Traditional manually operated gates often pose safety risks, increase security vulnerabilities, and require physical effort from users. This project presents the virtual design and simulation of an automated residential sliding gate using SolidWorks for mechanical modeling and Proteus for electronic control simulation. The system integrates key mechanical components such as the gate frame, rollers, track, and rack-and-pinion mechanism, alongside a microcontroller-based control circuit designed to operate the motor responsible for gate movement. The SolidWorks simulation was used to analyze the gate’s mechanical performance, focusing on linear motion, component alignment, and the conversion of rotational motor input into smooth sliding action. Proteus was employed to simulate the automation logic, including motor activation, direction control, and stopping at predefined limits. These simulations allowed full validation of system behavior without physical prototyping, reducing cost and eliminating real- world testing constraints. Results from both platforms confirmed that the gate moves smoothly, responds correctly to control inputs, and maintains proper synchronization between mechanical and electronic subsystems. The study demonstrates that virtual simulation tools provide an effective method for evaluating automated gate mechanisms before fabrication. The design also offers a foundation for future enhancements such as remote wireless control, improved safety features, and integration with smart-home systems.

An Automated Residential Gate project aims to enhance security, convenience, and energy efficiency through the integration of automation and solar power technology. Traditional manual gates require significant human effort and are often inconvenient, especially for large or heavy gates. To address these issues, this project involves designing an automated sliding gate system controlled by remote access, keypads, and IOT connectivity. The system incorporates a D5V6 Smart Centurion Machine, a 60W solar panel, a 30A charge controller, and a deep-cycle battery to ensure uninterrupted operation, even during power outages. The design includes a 0.37 kW motor with a gearbox to enhance torque efficiency, along with infrared sensors for obstacle detection and limit switches for precise movement control. Safety features such as emergency manual release and predictive maintenance alerts further improve usability and reliability. Structural materials such as steel and corrosion-resistant components ensure durability under various environmental conditions. Through performance testing, the system demonstrated smooth operation, energy efficiency, and enhanced security compared to conventional gates. The solar-powered system effectively reduces reliance on grid electricity, making it a cost-effective and sustainable solution. Future improvements may include AI-driven security enhancements and higher-efficiency solar panels to further optimize performance.

With increased urbanization and the growing need for better sanitation hygiene practices across several countries, including many developing nations such as Nigeria, there was a great need for better sanitary facilities that would allow for cleaner sanitation. Most conventional latrines have required high manual intervention and constant maintenance, creating an environment that is often dirty, wastes water, and does not have enough cleaning facilities, even in places like restrooms at homes and public places. Despite some innovative restrooms being designed before, the majority of such restrooms either lack feasibility financially or do not contribute effectively to smart hygiene monitoring. The smart toilet design mentioned in this paper seeks to overcome the above challenges. The proposed system comprises User Detection using Sensors, automatic Flushing, Optimization of Water Usage, Surface Hygiene monitoring, with Logic implemented using a microprocessor, and Wireless Communication as well. The design employs Infrared and Proximity Sensors for detecting the presence of the user and seat usage. The system utilizes solenoid valves for controlling the amount of water used in flushing. Furthermore, the proposed system will incorporate Adaptive Flushing Logic whereby the amount of water will be optimized to improve its usage and conserve more water. Testing will be done to evaluate the performance and ensure the proper functioning of the sensors and accuracy of the flushing. After completion of the project, the expected outcome would be a working smart toilet with automatic flush capability. Moreover, after testing it is expected that the system will be capable of reliably detecting users and communicating wirelessly while optimizing water usage as well as ensuring restroom hygiene. Not only is the project useful for home and public usage but also it will go a long way to improving the sanitation level in the ever-growing cities.

This research presents the design and implementation of a low-cost Artificial Intelligence-Based Battery Management System (AI-BMS) for lithium-ion batteries used in portable devices, solar power systems, and small electric vehicles. The study addresses thelimitations of existing systems-traditional thres hold based BMS that offer only reactive protection, and commercial smart BMS that are prohibitivelyexpensive and powerhungry.Using real degradation data from the NASA Prognostics Data Repository,an XGBoost machine learning model was developed to predict the State of Health (SoH) anddetect early thermal runaway precursors through voltage, current, and temperature trends. The trained model was deployed on an ESP32 microcontroller, integrated with low-cost sensors (INA3221 and ADS1115) and a 128×64 LCD for live system feedback.The AI-BMS achieved ±1.87% SoH accuracy, 100% safety response in fault simulations, and an average response time of 0.82 seconds, all at a total cost of approximately ₦22,450 ($18). Compared to conventional threshold-only protection and mid-range commercial BMS units, the proposed system offers proactive fault detection, predictive analytics, and real-time monitoring at a fraction of the cost and power consumption (35–48 mA). This study demonstrates that affordable, intelligent, and locally assembled BMS solutions can significantly enhance battery safety, extend lifespan, and democratize access to advanced energy storage technologies in developing regions.