Supervisor: Kofi Adu-Larbi

Project Description:
Disputed goals are a recurring problem in local football, often leading to unfair outcomes and reduced trust in officiating. Professional leagues rely on expensive goal-line technologies, but these systems are far beyond the reach of community football. GoalSense is a low-cost alternative designed to provide accurate, real-time detection of whether the ball has fully crossed the line. By focusing on affordability, accessibility, and ease of deployment, this project brings advanced sports fairness technology to grassroots leagues.

Supervisor: Acheampong Antwi Afari

Project Description:

Conventional vehicles rely on foot pedals for acceleration and braking; a system that immediately excludes individuals who cannot effectively use their legs due to stroke, accidents, or congenital disabilities. For people living with lower limb immobility, this presents a major barrier to independence, mobility, and quality of life.

Although hand-control systems exist on the market, they are often too expensive to import, leaving many local drivers to depend on unsafe, improvised solutions such as poorly fabricated sticks. Current systems force drivers to juggle the steering wheel, gears, and controls simultaneously, which increases fatigue and creates serious safety risks.

This project is centered on solving that gap. The goal is to design and prototype an affordable, ergonomic, and reliable hand-control mechanism that allows safe acceleration and braking using only the hands. By focusing on comfort, accessibility, and local adaptability, this assistive device will empower drivers with lower limb disabilities to enjoy safer, more inclusive, and more independent mobility.

Student: Yvette Ahishakiye

Supervisor: Eleazar Archer

Project Description:

In regions like Sub-Saharan Africa, where food insecurity remains a significant challenge, proper irrigation is critical for ensuring the early development and high yield of staple crops. Access to reliable moisture, especially during unpredictable dry spells, is essential for mitigating climate shocks and adapting to a changing environment. However, existing irrigation solutions face major adoption barriers. Drip irrigation, while effective, is a static and expensive system that requires extensive setup across an entire farm, making it economically unfeasible for many medium-to-large-scale farms. Similarly, conventional truss irrigation systems, which span large areas, are generally too bulky and static. Their immense size and weight make long-distance transport as a single piece of equipment impossible, limiting their feasibility and scalability.

This project seeks to address these critical gaps by developing an Origami-Inspired Foldable Truss for Mobile Irrigation. The core innovation lies in designing a system that combines the broad coverage of a truss with the portability of a compact device. By leveraging principles from biomimicry—specifically drawing insights from the structure of a bird's wing—and origami, the project aims to create a truss that can be folded into a compact form for easy long-distance transportation. This design approach will enable a single, highly mobile irrigation system to service large areas efficiently and quickly.

Supervisor: Prof Nathan Amanquah

LoRaWAN networks are privately owned but increasingly shared among different users and organizations. This raises critical challenges of trust and fair billing between network owners and users.

This project explores how blockchain technology can provide a decentralized, tamper-proof, and transparent framework for managing LoRaWAN networks. By integrating blockchain smart contracts, the system ensures that:

• Network usage is securely tracked.

• Owners can bill users fairly.

• Participants benefit from a trusted, decentralized infrastructure for IoT connectivity.

Supervisor: Richard Awingot Akparibo

Project Description:

Current renewable hot water solutions are expensive - solar PV + battery systems cost about $10,000+ with decade-long battery replacements, while solar thermal requires complex plumbing and freeze protection. This capstone project explores whether combining solar PV with fine sand thermal storage could provide a low-cost, environmentally friendly alternative to conventional electric geysers by using abundant, inexpensive sand to store solar energy as heat instead of expensive batteries. Through component research, system design, and prototype testing, the project investigates the technical feasibility of this integration for reliable domestic hot water applications. This research isn't just about hot water - it's about exploring whether we can democratize renewable energy storage. If successful, this technology could potentially transform how homes store solar energy, making renewable living affordable and accessible worldwide.

Supervisor: Alhassan Sullaiman

Project Description:

Household drainage systems in kitchens and bathrooms frequently suffer from buildups of grease, soap scum, hair, and oils, leading to slow drainage, foul odors, and eventual clogging. Current solutions, such as chemical cleaners, risk damaging pipes and harming the environment, while professional plumbing services are costly and typically reactive rather than preventive. In large households and multi-unit buildings, the complexity of pipe networks makes it even harder to identify blockages and maintain the entire system effectively.

This project proposes the design of a smart, compact pipeline PIG for household drainage cleaning and maintenance. The PIG will feature a helical vane design that spins with water flow to scrub pipe walls, while battery-assisted motors provide extra torque when needed. A tracking and localization system will enable users to monitor its position, ensuring recovery if stuck, offering a reusable, environmentally friendly, and cost-effective solution for preventing clogs in household pipelines.

Supervisor: William Akuffo

Project Description: Many prosthetic hands often struggle to replicate the dexterity of biological hands with many commercial models having only up to 10 DOFs whereas the human hand has about 27 - a vast difference, especially, for forearm amputees who once had fully dexterity and autonomy of their hands. These amputees retain neural pathways and muscle groups in their arms that could be used to detect and identify intended motion. The capstone project seeks to model, study and identify the possibility of using these neuromuscular inputs to control and improve the dexterity of prosthetic hands hopefully with the aim of making improved dexterity more commercially available.

Supervisor - Dr Stephen Kofi Armah

This project focuses on the design, analysis and fabrication of a dual-axis solar tracking system that enhances the efficiency of photovoltaic panels. An application of machine design, manufacturing processes, embedded technology, control system and automation, among other engineering principles will be employed. The technical aspect of the project is to continuously align the panel with the sun’s position in both azimuth and elevation.

Supervisor: Elena Rosca.

Technical Supervisor: Lloyd Teta.

Project Description:

We are developing a bio-inspired ornithopter, a lightweight, flapping-wing aircraft designed to mimic bird flight, for discreet aerial surveillance in Ghana’s Galamsey (illegal mining) regions. Unlike noisy drones, the ornithopter can blend into natural environments, allowing it to monitor deforestation, water pollution, land degradation, and even wildlife with minimal disturbance.

Equipped with a camera system, wireless control, and an energy-efficient frame, this project not only supports environmental protection but also contributes to the fight against illegal mining while promoting sustainable development goals (SDG 15 & SDG 16).

Supervisor: Eleazar Archer

Project Description:

Currently, the primary method for producing plastic components is crude 3D printing. While useful for rapid prototyping, this process severely limits a part's strength, durability, and the potential for manufacturing process scaling for both batch and unit volume builds. This reliance on a single, limited process often results in projects that are discontinued rather than developed into full-fledged, market-ready products.

To address these significant limitations, this project aims to develop a Knowledge-Based Engineering (KBE) system and a process flow as a solution. This system will serve as a formalized knowledge base designed to guide users through the complex process of designing parts for carbon fiber manufacturing and their corresponding molds. 

By capturing and institutionalizing expert knowledge into a computational framework, this project seeks to establish a more efficient, reliable, and accessible method for composite design and manufacturing on campus, empowering students and faculty with varying levels of expertise to create truly functional and scalable products.

Supervisor: Bright Tetteh

Project Description:

This project focuses on creating a power electronic interface that allows energy to flow in both directions between a battery and a DC source. In charging mode, the converter steps down the DC bus voltage to safely charge the battery, while in discharging mode, it steps up the battery voltage to supply power back to the system. Such converters are essential in renewable energy integration, microgrids, and backup systems, ensuring efficient energy storage and reliable power supply.

Student: Jada Abena Animwah Boateng

Project description:

Capstone Project Description

The global electric vehicle market has experienced rapid growth since 2017, with over 58 million vehicles on the roads worldwide as of 2024. The primary markets driving this expansion are China, Europe, and the United States. However, as demand for electric vehicles continues to rise in alignment with Sustainable Development Goal 12 (responsible consumption and production), several technical challenges have emerged, particularly noise and vibration issues that significantly impact both battery performance and passenger comfort.

This project aims to analyze how the orientation and geometry of lithium-ion batteries affect the charging and discharging rates, as well as vibration characteristics in electric vehicles. The research will involve a comprehensive computational fluid dynamics (CFD) analysis to examine the thermal effects of various battery orientations (vertical, horizontal, and inverted) and different shapes (cylindrical and prismatic) across different frequency ranges. This analysis will help identify the mechanical stresses that influence battery degradation and overall performance.

Project Description:

"Globally, about 1/3 of food is lost or wasted, which results in an economic loss of about $1 trillion a year. If food waste and loss were a country, it'd be the third-largest emitter after China and the United States" (World Economic Forum,2022).

Food spoilage has become a persistent challenge in commercial storage facilities, causing substantial financial losses, food insecurity, and health risks. Perishable goods like fruits, vegetables, and dairy often deteriorate faster due to hidden changes in factors like temperature and gas emissions. As a result, tons of food are wasted, significantly impacting operational costs. In cases like these, spoilage goes unnoticed until it is too late.

Addressing the challenge requires timely detection and effective management to minimise spoilage and ensure food safety.

This capstone project proposes a system that not only detects spoilage through multiple monitoring methods but also anticipates the risk of spoilage via predictive schemes. Facility managers would be able to view risk levels, receive real-time alerts, and access spoilage scores that indicate how conditions and factors of spoilage are changing over time, while having minimal reliance on internet connectivity, with support features for ventilation regulation.

Supervisor: Kofi Adu-Labi

The maintenance of large grassy fields remains a resource-intensive activity for schools, sports facilities, and public institutions. Conventional mowing methods rely on manual labor and fuel-powered machines, both of which present significant challenges. Manual mowing is physically demanding, generates considerable dust that poses respiratory health risks, and requires extended working hours. Fuel-powered machines, while effective, contribute to air and noise pollution, greenhouse gas emissions, and rising operational costs associated with fuel consumption.

Advances in electric drive systems, battery storage, and autonomous navigation technologies provide a basis for developing more sustainable approaches to field maintenance. Existing autonomous mowers, however, are predominantly designed for small residential lawns and are not adapted for larger areas such as football pitches.

This project proposes the design and prototyping of an electric autonomous mower specifically for wide fields, with initial application on the Ashesi University football pitch. By integrating electric propulsion, sensor-based navigation, and intelligent control, the system seeks to deliver a reliable, scalable, and environmentally responsible alternative to conventional mowing practices.

Supervisor: Dr. Nathan Amanquah

Modern vehicles are increasingly equipped with electronic control units (ECUs) and sensors that generate vast amounts of data. This data is transmitted over the Controller Area Network (CAN) bus  and provides insights into driving behavior, vehicle performance, and maintenance needs. At the same time, the demand for systems that can monitor and evaluate driver habits has grown rapidly, driven by applications in usage-based insurance, fleet management, fuel economy optimization, and predictive maintenance. Integrating embedded machine learning (ML) with IoT-enabled systems makes it is possible to create intelligent solutions that go beyond basic data logging and provide real-time analysis and actionable feedback.

This project proposes the design and implementation of a Driving Habits Monitor that leverages embedded ML algorithms and CAN bus interfacing. The system will collect key vehicle parameters (such as speed, braking intensity, acceleration, and engine performance) directly from the CAN bus. Using embedded ML, it will classify and analyze driving patterns to distinguish between safe, aggressive, or inefficient driving styles. Additionally, environmental impacts on the vehicle such as frequent exposure to rough terrain, bumps, or dust will be monitored to provide maintenance recommendations.

The system will also feature IoT capabilities to track vehicle location and transmit driving reports to remote servers or cloud platforms for further analysis and record-keeping. This makes it highly useful for insurance companies evaluating risk profiles, for individuals aiming to improve fuel economy, and for fleet operators monitoring driver compliance and vehicle health.

Supervisor: Dr Stephen Kofi Armah

This project focuses on the design, analysis and fabrication of a dual-axis solar tracking system that enhances the efficiency of photovoltaic panels. An application of machine design, manufacturing processes, embedded technology, control system and automation, among other engineering principles will be employed. The technical aspect of the project is to continuously align the panel with the sun’s position in both azimuth and elevation.

Supervisor: Professor Robert Sowah

Cocoa quality , which determines the market value of its bi-products , depends heavily on post-harvest processes such as fermentation and sorting. For smallholder farmers in Ghana , these processes are highly manual and inconsistent which accounts for post-harvest losses of up to 20%. Traditional methods lack precision ,while advanced alternatives remain costly and inaccessible. This project proposes Invisi, an intelligent IoT based system that integrates AI-enabled fermentation pods with a computer vision sorter to deliver consistent cocoa processing. The fermentation pods autonomously regulate and measure temperature , humidity and gas levels for optimal flavor development, while the sorter classifies beans into high quality, defective or under fermented categories. Both components generate digital quality reports that inform the farmer of their stock quality and issues that need to be resolved such as fermenting some beans a while longer. The project aims to deliver a functional system , machine learning models and a farmer dashboard. By having an autonomous consistent system Invisi hopes to create a consistent bean quality and reduce the post-harvest losses due to inefficiencies and inconsistencies in the initial processing phase of cocoa beans.

Supervisor: ........

Hypertension and diabetes affect over 20% of adults in Ghana, yet awareness, consistent monitoring, and early intervention remain low. This has contributed to stroke becoming one of the leading causes of death and long-term disability in the country. Many cases are linked to undiagnosed atrial fibrillation (AFib) and poorly managed blood pressure. While medical devices exist to monitor these conditions, they are often too expensive, too complex for everyday use, or fail to translate clinical measurements into clear, actionable feedback that ordinary people can understand and act upon.

This project introduces a portable point-of-care device that integrates blood pressure, heart rate, ECG, and SpO₂ sensors into a single, affordable solution. Using AI models trained on both population datasets and individual patient patterns, the device can screen for AFib, estimate stroke risk over the short and medium term, and then translate the results into plain, easy-to-understand language. Patients receive clear insights and lifestyle recommendations, such as diet adjustments, activity reminders, and clinic visit prompts, while clinicians receive structured reports that can be integrated into existing hospital systems.

Designed to be low-cost, portable, and usable even without internet connectivity, this solution empowers Ghanaians, especially those in rural and underserved areas, to take control of their health before emergencies occur. By detecting risks earlier and encouraging preventive action, it has the potential to reduce stroke-related deaths, disabilities, and the long-term costs of chronic disease management.

Supervisor: William Akuffo

Project Description.

Background

Modern educational platforms like the inverted pendulum, Q3Bot, and QBot, produced by companies such as Quanser, have proven essential for research and teaching purposes in robotics and mechatronics. However, their exorbitant costs, ranging from $3,000 to $50,000 pose a significant barrier for universities in developing countries. This financial limitation forces many institutions to rely solely on simulation software such as MATLAB and ROS, which, while effective, deprive students of the hands-on experience necessary to develop practical skills.

Objective

This project aims to bridge this educational gap by developing a low-cost tail-sitter Bi-Copter platform. The platform will serve as an affordable and locally accessible tool to enhance the practical teaching and learning process in mechatronics, robotics, and engineering disciplines.

Supervisor: Kofi Adu-Labi

Poultry farming plays a vital role in ensuring food security and strengthening Ghana’s economy, contributing about 14% to the country’s GDP. Despite the rising demand for poultry meat and related products, local production continues to fall short of meeting the needs of the growing population. One of the major setbacks facing the industry is the frequent outbreak of diseases.

For instance, in 2015 alone, over 703,000 birds were lost to a Highly Pathogenic Avian Influenza (HPAI) outbreak. Newcastle disease remains another persistent threat, causing significant losses on many farms. Currently, most farmers rely on manual inspection to detect sick birds, a method that is inefficient, labor-intensive, and highly unreliable, especially for medium to large-scale farms.

This project seeks to address this challenge by introducing an IoT- and machine learning–based solution. Using a combination of infrared sensors, cameras, and microphones, the system continuously monitors bird movement, sound, and temperature. Machine learning models then analyze the data in real time to detect abnormal patterns, allowing farmers to receive timely alerts and respond quickly to potential disease outbreaks.

Supervisor: Dr. Elena Rosca

Project Description:

Airborne pathogens remain among the most critical threats to indoor safety, as highlighted by the COVID-19 pandemic, which claimed 7,101,631 deaths worldwide (World Health Organization, 2025). But what if there was a way to constantly protect ourselves from these airborne pathogens so that we cannot succumb to any diseases and remain free from any airborne diseases. Because current air purifiers only remove particles but lack intelligence: they cannot detect what pathogens are present, cannot remember exposures, and cannot adapt. This project develops a prototype artificial indoor immune system that mimics how the human immune system works. 

So, just as the human immune system has an innate immune system that produces rapid but non-specific responses to pathogens, this project will have a system that detects the pathogens in the air and processes the signals in a matter of seconds to give a fast, non-specific defense mechanism to the building, protecting it from the airborne pathogens. The human immune system also has an adaptive immunity system, which is composed of highly specialised cells (such as the T and B cells) and processes that eliminate specific pathogens. This project will adopt that same analogy and take the same aerosol liquid sample used as an innate response and pass it through a cell-free system, which will then ensure that it processes the DNA of the pathogens to give a specific defense mechanism for that particular pathogen. In the same way, the human adaptive immune system has an immunological memory. This project will ensure that when the cell-free system identifies the pathogens, the memory of that pathogen will be encoded for later quick response of the system, and to also inform us what kind of pathogen was detected and at what time, and the defense mechanism that was used.

The aim is to build a living, learning, and reacting artificial immune system for homes that goes far beyond traditional purification.

References

World Health Organization. (2025). COVID-19 deaths | WHO COVID-19 dashboard. World Health Organization Data. https://data.who.int/dashboards/covid19/deaths

‌

Supervisor: Dr. Stephen Armah

Project Description:

As the global demand for renewable energy grows, solar power has become one of the most practical and widely adopted options. However, a surprisingly common problem holds back its full potential - the build up of dust and other forms of dirt on the panels. Over time, this layer of grime can block sunlight and cut energy output by as much as 30%, especially in dry or dusty regions. Keeping panels clean is essential, yet the usual approach of sending people out to clean them is costly, time-consuming, and often impractical for installations in remote areas.

We plan to address this issue by developing a Smart Solar Panel Cleaning System that combines sensors, machine-learning based vision analysis, and automated cleaning. The system would be designed to monitor panels, detect when dust and debris begins to affect performance, and then carry out cleaning routines on its own, without interrupting energy generation.

To make the system even more practical, we might add an IoT platform that lets users check panel status in real time, view performance trends, and schedule cleanings if they wish to do so. By shifting from manual, reactive cleaning to a smart, data-driven approach, we aim to help solar installations stay efficient, reduce maintenance costs, and keep the transition to renewable energy moving forward.

Supervisor: Dr Stephen Armah

Project Description:

Existing wingtip devices on commercial aircraft are tuned to specific flight conditions (takeoff, climb, or cruise), so their effectiveness drops as flight conditions change, limiting drag-reduction benefits and offering little gust-load relief, further worsening ride quality. This project seeks to develop an adaptive wingtip device that detects operating state and turbulence in real time and adjusts accordingly to sustain aerodynamic efficiency and alleviate gust loads.

Student: Kaare Yirviel Tietaah
Supervisor: Awingot Richard Akparibo

Solar Energy harnessed by Photovoltaic (PV) panels is, and has been, the leader in the vast adoption of renewable energy generation options (contributing about 81% of the world's renewable energy), in a global effort to move towards more sustainable energy methods and decarbonization (SolarPower Europe (2025): Global Market Outlook for Solar Power 2025-2029). Solar PV power generation is variable and sensitive to environmental conditions such as solar irradiance and temperature. Maximum Power Point Tracking (MPPT) algorithms  are needed to maximize the  PV output power. Conventional solar PV controllers, however, typically act as black-box optimizers. This means they do not expose the panel’s full electrical behaviour and cannot reliably detect faults such as partial shading, soiling, cell mismatch, open circuits, etc. until energy production declines noticeably. Offline I-V tracers and laboratory equipment provide complete characterization of panels, but are disruptive, costly, and unsuitable for real-time continuous field monitoring.

This project proposes TRACE: an embedded DC–DC converter that doubles as a measurement instrument. While tracking the maximum power point in real time, TRACE performs low-intrusion micro-probing and occasional implementing duty-cycle sweeps to reconstruct the panel’s I–V / P–V behaviour on the fly. It then fits a compact physics-based PV model and uses those parameters to detect and classify faults before they erode yield. The hybrid approach balances energy capture with continuous health monitoring, turning the controller from a passive optimizer into an active diagnostic engine.

Beyond improving day-to-day efficiency, TRACE enables actionable maintenance: early alerts for shading or soiling, flags for connection faults or emerging hot-spots, and trend data for lifecycle planning. With the aims of being cost effective and field-deployable, the system targets high tracking efficiency, making advanced PV analytics feasible at the module edge, where reliability, uptime, and every Watt matter most.

Supervisor: William Akuffo

Project Description:

This project focuses on the development of a versatile 6-axis robotic arm platform designed to support both teaching and research in robotics and control. Unlike industrial-grade robotic arms that are often costly and limited in accessibility, this platform emphasizes affordability, modularity, and adaptability, making it ideal for academic environments. The system is intended to serve as a hands-on tool for students to learn fundamental concepts such as kinematics, dynamics, trajectory planning, and control strategies, while also offering researchers a flexible testbed for advanced studies in automation, artificial intelligence, and human-robot interaction. With an open-architecture design and compatibility with widely used hardware and software frameworks, the platform promotes experimentation, customization, and collaboration. By bridging the gap between theoretical learning and practical implementation, this project aims to enhance the educational experience in robotics while enabling innovation and exploration in research. Ultimately, the proposed robotic arm platform seeks to democratize access to robotics education and provide a scalable solution that can evolve with the growing demands of the field.

Supervisor: Acheampong Antwi Afari

This project aims to improve a semi-automatic screw-press briquette machine that uses sustainable raw materials such as coconut fibre, rice husk, and cassava starch to produce charcoal. The current design produces uneven briquettes due to manual cutting and uses fixed input power, limiting quality. Improvements include adding an automated cutter for uniform size and density, testing variable input power to optimize properties, and analyzing the enhanced briquettes against existing ones.

Supervisor: Alhassan Sullaiman

Project Description: Post-harvest losses remain a major challenge in agriculture, with perishable crops such as fruits and vegetables wasted due to inadequate storage and handling. In many local markets, significant losses occur because of poor logistics; delays in transportation, lack of proper cold chain facilities, and insufficient storage capacity. Existing cold storage systems are energy-intensive, poorly optimized for airflow, and fail to meet the diverse temperature and humidity requirements of different crops. This results in uneven cooling, condensation, and accelerated spoilage, particularly in resource-constrained environments.

This project seeks to design and simulate an optimized multi-crop storage container that addresses these challenges. The design will integrate both passive and active cooling methods, such as evaporative cooling and mechanical refrigeration, to maintain crop-specific preservation conditions. The study will analyze airflow distribution, duct design, tray spacing, and heat transfer to maximize efficiency and minimize energy losses. The project will also explore sustainable insulation materials, such as chicken feathers, and apply psychrometric analysis for effective humidity control.

Supervisors: Dr. Yaw Delali Bensah & Dr. Stephen Armah

Project Description:

The increasing use of fossil fuels in combustion engines continues to drive greenhouse gas emissions, climate change, and local air pollution while creating economic vulnerability for developing regions like Ghana that rely heavily on petroleum imports. Palm kernel oil, an abundant and underutilized agricultural by-product in Sub-Saharan Africa, offers a promising feedstock for sustainable biodiesel production. However, current small- to medium-scale systems struggle with inconsistent fuel quality, inefficient conversion, and limited process control.

This project develops a modular biodiesel production unit optimized for ethyl ester synthesis from palm kernel oil. The plant integrates crushing, oil extraction, and transesterification processes in a skid-mounted design for scalability and portability. CAD-based design and analysis will be applied to critical subsystems such as the kernel crusher, oil reactor, and purification modules, with stress, vibration, and thermal performance considered. Industrial sensors integrated with a PLC-controlled automation system will enable real-time monitoring, safety interlocks, and process optimization. The overarching goal is to optimize reaction conditions to deliver high-yield, high-quality biodiesel while enhancing energy efficiency, operational safety, and environmental sustainability, ultimately contributing to cleaner energy transitions and value creation from local resources.

Supervisor: Dr. Elena Rosca

Bacterial cultivation is at the heart of biotechnology, medicine and research, yet many low-cost cultivation systems still leave users working working harder and not ultimately smarter. Most bench-top bioreactors only monitor basic parameters like temperature and pH, without providing an autonomous way to identify the most productive stage of bacterial growth: the log phase.

This project aims to change that by developing an IoT-integrated bioreactor that not only measures conditions but interprets them in real time, offering a smarter and more accessible way to cultivate microbes. At its core is a turbidity sensor module, which uses light scattering technology to estimate optical density and, in turn, cell count. Paired with precise temperature and pH regulation mechanisms, the system ensures that bacteria grow consistently in optimal conditions.

The true innovation lies in the module and algorithm that would autonomously detect the onset of the log phase by analyzing turbidity trends over time, eliminating the need for constant manual checks. All this data feeds into an IoT dashboard for remote monitoring, visualization, and instant alerts, empowering users to make timely and informed decisions.

Supervisor:
Dr. Nathan Amanquah

Project Description:
Embedded and edge machine learning systems face tight constraints in speed, power, and scalability. Previous work demonstrated that FPGA-based softcore processors with on-chip accelerators can achieve significant latency improvements compared to microcontroller-based setups. However, questions remain about scalability, power efficiency, and how FPGA-based acceleration compares with mainstream GPU-based acceleration.
This project will build upon that foundation by benchmarking FPGA and GPU platforms across different neural network models (MLP, CNN), evaluating inference latency, power consumption, and efficiency. The study aims to highlight the tradeoffs between FPGA and GPU acceleration for embedded ML, providing guidance for real-world applications such as IoT, robotics, and edge AI.

Supervisor: Dr. Stephen Kofi Armah

The transport market is the second largest energy consumer in Ghana, accounting for 34% of total final consumption in 2019, and was responsible for 17% of national GHG emissions in 2016. Road transportation, particularly passenger vehicles, heavy-duty trucks, and buses, is the largest source of these emissions. 

To align with 1.5°C-compatible pathways, Ghana must reduce direct CO₂ emissions from transport. One way to do so is to improve the energy efficiency of existing road transport, particularly for freight and export logistics, as road transport is the dominant mode of moving agricultural products for export in Ghana.

This project proposes to design and test modular aerodynamic add-ons to reduce drag and improve fuel efficiency in heavy-duty trucks operating on Ghanaian highways. An active control element, responsive to vehicle speed and ambient wind yaw angles, will be incorporated to dynamically adjust aerodynamic surfaces to minimise drag coefficient increments with yaw angle.

Supervisor: Dr. Nathan Amanquah

GPS-Based Smart Traffic Light Control System

Project Description:

This project proposes the design and implementation of a smart traffic light system that uses Global Positioning System (GPS) data from vehicles to optimize signal timing in real time. Vehicles within the network periodically transmit their location, speed, and direction to roadside units (RSUs) installed near intersections. Using this information, each intersection controller predicts the estimated time of arrival (ETA) of approaching vehicles, analyzes traffic density, and dynamically adjusts green, yellow, and red phases to reduce congestion and waiting time.

The system prioritizes emergency vehicles by instantly detecting their presence and computing the fastest route for safe and immediate lane clearance. Unlike traditional fixed-timing signals, this GPS-driven approach allows adaptive control based on actual vehicle movement rather than pre-set schedules.

Supervisor: Acheampong Antwi Afari

Project Description:

This project focuses on the design and development of an automated beach cleaner to address the challenge of pollution caused by human activity along coastlines. Beaches play a vital role in tourism, recreation, and marine ecosystems, yet they are increasingly threatened by plastics, cans, and other waste that are difficult to remove manually. The proposed device, inspired by the functionality of a lawn mower, will be designed to scoop and filter solid waste from sand while separating it from the natural environment. By automating this process, the project seeks to reduce the heavy human effort and time required for beach maintenance while offering a sustainable and environmentally friendly approach to conservation. Ultimately, the beach cleaner aims to promote sustainable tourism, and contribute to marine protection by ensuring cleaner and healthier coastlines.

Student: Calvin Mac Phillips

Supervisor: Dr. Patrick Dwomfuor

Project Overview

Ghana faces a significant waste-management challenge. Large amounts of refuse continue to degrade land, pollute water bodies, and harm the atmosphere. Traditional responses—such as monthly communal clean-up exercises—have proven inadequate for tackling the scale and complexity of the problem.

This project proposes a proof-of-concept smart bin that automatically sorts household waste common in Ghana. The system leverages a locally trained computer vision model to identify and separate different waste types, promoting more efficient recycling and reducing environmental impact. Some key objectives are to evaluate the economic and environmental viability of deploying low-cost smart bins in Ghanaian homes, investigate the trade-off between classification accuracy and inference speed, and ensure the model runs efficiently on affordable hardware. Lastly, compare candidate computer vision architectures to determine the best fit for real-time waste sorting under local conditions.

Supervisor: Dr Elena Rosca

Project description:

In many parts of the country, especially in rural areas, girls are not prepared and aware of menstruation, so they face many difficulties at home and in schools. Some women do not have access to sanitary products, and sometimes menstrual cycles can be unpredictable; having access to sanitary pads at the right time is crucial. This project addresses these challenges by providing easy and affordable access to sanitary products through a solar powered system with stock prediction capabilities.

Supervisor: Acheampong Antwi Afari

Project Description:

Background:

Second hand trade is essentially the backbone of many women and families in Accra that trade in the market. A recurring challenge that they face is buying damaged bales which means they encounter financial losses. There is no affordable or practical way for the market women to dispose of the damaged clothes. Therefore, torn and damaged clothes are dumped in landfills and beaches polluting the environment and making beaches in Accra uninhabitable.

Objective

The aim of this project is to build a low-cost, human-powered mechanical shredder that can provide a sustainable solution. By shredding the torn fabrics into smaller pieces, the material can be repurposed for various applications such as cushion filling, cleaning rags, insulation, or even raw materials for small-scale recycling. This not only reduces environmental pollution but also provides market women with an opportunity to turn waste into value, thereby improving their livelihoods.

Supervisors: Dr. Yaw Delali Bensah & Dr. Miriam Abade-Abugre

Project description:

In Ghana, two pressing issues stand out in construction and waste management. First, millions of bottles and windows made from soda-lime glass end up in landfills and open dumps because there is no large-scale recycling system. Second, cement blocks widely used in building construction often absorb water, leading to cracks and, in worst cases, structural collapse.

My project seeks to tackle both problems by developing foam glass from recycled soda bottles and window glass and transforming it into lightweight, water-resistant building blocks. By experimenting with glass foaming techniques and blending the foam glass with a controlled amount of cement, the goal is to create blocks that:

• resist water absorption (reducing cracks and failures),

• are strong enough for building use, and

• provide a sustainable alternative to conventional cement blocks.

This innovation not only addresses a critical safety issue in Ghana’s construction industry, but also creates a circular economy pathway for glass waste, turning an environmental challenge into a building solution.

Cocoa is one of the major exports of many West African countries, predominantly Côte d'Ivoire and Ghana, with a large proportion of cocoa produced by small-scale farmers. However, due to the large size of the average plantation, it is difficult for small-scale farmers to monitor the farm and identify diseases in the plants before they spread. Diseases alone are estimated to cause about 30 – 40% of the losses in cocoa production. [1].

Building upon previously done ground robot projects, this capstone project aims to improve upon the navigation capabilities, specifically the path planning and obstacle avoidance capabilities, of the ground robot, enabling it to manoeuvre semi-autonomously around the plant to monitor the trees as well as detect some disease that can cause significant losses in plants like Black Pod and Cocoa Swollen Shoot Virus Disease (CSSVD). This capstone also seeks to integrate a communication protocol and web interface to make this system more user-friendly.

Problem

Improve navigation, implement communication protocols and a web interface to improve ease of use for farmers, and implement disease detection to reduce the chances of widespread disease and high losses.

Objectives

·       Enhance Path Planning and Obstacle Avoidance: Implement obstacle avoidance algorithms and research into path planning algorithms to determine which algorithm or combination of them will best suit the terrain of cocoa farms.

·       Integrate Communication protocols and implement Web Interface: integrate communication protocols to allow the robot to communicate with any other sensors on the farm and with a device that contains a web interface that would enable the farmer to communicate with and direct the robot easily.

·       Implement disease detection algorithms: Research into and implement an algorithm that uses data from cameras and sensors on the robot to identify which plants are healthy and which ones may be diseased, as well as what diseases the plant may have. This will then direct the farmer to what resources can be used to confirm the disease and how to prevent it from spreading while waiting for confirmation.

Supervisor: William Akuffo

Project Description:

There is no doubt that with an increase in urbanization in developing countries, there is a reduction in population in rural communities. These communities are known to be one of the main providers of food in the country, and hence, this shift in population has created labor shortages and resource constraints in the agricultural sector. Precision Agriculture (PA) offers a solution by introducing smart farming, which utilises technology in robotics and automation to perform essential farming tasks more efficiently.

To understand the potential of such systems, it is vital to note that crops grown on a field vary in terms of growth time and amount of space occupied. Although there have been various implementations of PA, most rely on aerial systems and GPS-based solutions, which are not always effective in fields with crops that have complex spatial and temporal requirements.

As such, the aim of this project is to address this gap by developing a mobile ground rover equipped with a manipulator and guided by the Extended Kalman Filter (EKF)-based SLAM (Simultaneous Localisation and Mapping) concept. The system will enable autonomous navigation and monitoring of crop fields, with a focus on capturing variations in space and time to support more precise and sustainable farm management.

Emmanuel Bayour Basakeng

Project Description:

Industrial equipment relies heavily on effective maintenance strategies to prevent costly downtime and safety hazards. Traditional approaches, which are reactive maintenance (fix-when-broken) and scheduled maintenance (fix after a set time), often result in operational inefficiencies, unexpected failures, or unnecessary production losses. Current predictive maintenance solutions typically monitor single parameters, limiting their diagnostic sophistication.

This project develops an intelligent predictive maintenance system that creates a digital twin of industrial equipment, continuously updated with performance, maintenance, and health status data. The system monitors machine signatures through multiple parameters, primarily vibration, acoustic, and temperature monitoring, to accurately identify specific component wear and predict failures before they occur.

Targeting critical machinery in mining, oil and gas, and chemical production sectors, including mud pumps, hoist systems, swing motors in excavators, CNC spindle motors, compressors, and generator bearings, the system employs a three-layer architecture: device layer (data collection via self-powered sensor modules), fog layer (real-time data processing using LSTM autoencoders), and cloud layer (data storage and advanced analytics).

The solution provides operators with a comprehensive dashboard displaying real-time equipment health metrics, component-specific wear levels, and predictive insights. By identifying which parts need replacement and when, this system enables proactive maintenance scheduling, reducing downtime, extending equipment lifespan, and improving operational safety and efficiency across high-stakes industrial environments.

Supervisor: Dr Stephen Armah

Project Description

Poultry production in Ghana is constrained by high chick mortality rates, particularly during the brooding stage, where young chicks require tightly regulated environmental conditions. Current practices often rely on basic charcoal or electric brooders, which provide limited control over temperature, humidity, and ventilation, and require continuous manual monitoring. These systems are highly vulnerable to erratic power supply and increasingly variable climatic conditions, resulting in significant losses that undermine farmer profitability and national food security.

Although automated brooding systems exist internationally, they are typically expensive, difficult to maintain locally, and not designed to withstand Ghana’s dusty and humid farm environments. This creates a gap in the availability of affordable and adaptable technologies capable of supporting farmers in managing critical brooding conditions under local constraints.

This project seeks to investigate how embedded systems, sensor technologies, and resilient power architectures can be applied to develop a brooding system that is both adaptive and locally viable. The research will focus on identifying design approaches that could enable automated climate regulation, maintain functionality during power interruptions, and provide timely feedback mechanisms to farmers. By addressing these gaps, the project aims to outline a pathway toward reducing chick mortality while enhancing the resilience and sustainability of poultry farming in Ghana.

In Ghana, two pressing issues stand out in construction and waste management. First, millions of bottles and windows made from soda-lime glass end up in landfills and open dumps because there is no large-scale recycling system. Second, cementitious and clay-based blocks/bricks widely used for building and construction often absorb water, leading to cracks, material disintegration and corrosion, and, in worst cases, structural collapse.

This project seeks to tackle both problems by developing a ceramic-foam based material from recycled soda glass. The goal is to transform waste soda glass into durable, strong, lightweight, and water-resistant building material. By experimenting with different glass foaming agents and techniques, the resulting material would be batched with a controlled amount of cement, with the aim of creating blocks that:

•           can resist water absorption (reducing cracks and failures),

•           are strong enough for building use, and

•           provide a sustainable alternative to conventional cement blocks.

This innovation not only addresses a critical safety issue in Ghana’s construction industry but also creates a circular economy pathway for glass waste and turning an environmental challenge into a building solution.

Supervisor: Acheampong Antwi Afari 

Project Description: 

Global electricity consumption has surpassed 30,000 terawatt-hours (TWh) annually, with continuous growth driven by transportation systems, household appliances, industrial operations, and digital infrastructure, which rely on either non-renewable energy, which can deplete or renewable energy, which depends on weather conditions, toxic emissions, infrastructure and cost. To help address energy demands and reduce power shortages, sustainable and innovative energy harvesting methods are needed. The aim of this project is to develop and optimise a low-cost, high-durability piezoelectric floor tile for sustainable urban energy harvesting with an efficiency of about 60%. The research will explore the factors of efficiency, redesign power transmission, select materials and iterate parameters through a series of calculations and simulations to power areas of public engagement like city centres and shared social amenities like Airports, Stadiums and Bus/Train stations

Supervisor: Kofi Adu Larbi

Project Description:

This project focuses on the design and development of a low-cost grip strength monitoring device that uses piezoelectric sensing and embedded progress tracking to support stroke and paralysis rehabilitation. The device is aimed at assisting recovering patients in remote or underserved areas who may not have easy access to premium medical care, regular physiotherapy, or frequent clinical checkups.

By capturing grip force through piezoelectric sensors and recording performance over time, the system provides a reliable and affordable means for patients to monitor their recovery progress from home. This empowers patients and caregivers to detect early signs of improvement or potential health concerns, enabling timely medical attention while reducing dependence on costly hospital visits.

Ultimately, the device bridges the gap between professional rehabilitation tools and accessible, low-resource healthcare solutions, offering a practical pathway for patients to regain independence and improve their quality of life.

Supervisor: Master Bright Tetteh.

This project seeks to design and implement an Advanced Home Energy Management System (HEMS) using smart sockets for real-time monitoring and control of household appliances. Each socket will integrate sensors and a microcontroller to track energy consumption, detect appliance states, and communicate with a central controller. Through a simple user interface, households will be able to monitor usage, remotely switch appliances on or off, and receive alerts for abnormal consumption or overload conditions. The system will also support automatic load prioritization during periods of limited power supply, running in parallel with a broader load management framework to ensure that critical devices remain powered while reducing waste from non-essential loads. By combining smart sockets with intelligent load allocation, this project offers a practical solution to energy inefficiency and unreliable supply in Ghanaian homes, ultimately helping users save power, extend backup runtime during outages, and improve household energy awareness

This project seeks to revolutionize building health monitoring by designing and developing an advanced fiber optic sensor system. This innovative sensor employs an optical fiber connected to a transducer that accurately converts strain and vibration-induced optical variations into measurable electrical signals. We will rigorously test the system on building structural elements under controlled loads to assess its sensitivity, accuracy, and reliability compared to conventional monitoring techniques.

The anticipated result is a cutting-edge prototype capable of detecting early signs of structural weakness and providing continuous, immune-to-noise monitoring. This groundbreaking work will significantly enhance the safety and longevity of buildings, marking a significant advancement in structural health monitoring and maintenance practices. This work contributes to advancing cost-effective structural health monitoring technologies and has potential applications in enhancing building safety, guiding maintenance, and extending the life of building structures.

Supervisor: Dr. Stephen Armah

This project focuses on the design and development of a semi-automated, solar-powered robotic system for debris removal in urban drainage channels in Ghana. The system will integrate mechanical collection mechanisms, embedded control, and solar-based power management to provide a sustainable and adaptable solution for clogged gutters and drains.

Equipped with mobility features, sensors, and semi-autonomous navigation, the robot will be capable of maneuvering through drainage pathways to collect and store a variety of debris. Computer vision and machine learning techniques will be employed to detect, classify, and track debris as well as to enhance navigation and decision-making in complex drainage environments.

The design emphasizes affordability, modularity, and local manufacturability to ensure practical deployment in Ghanaian urban settings. By reducing debris-induced flooding and improving waste management efficiency, the system will serve as a scalable technology for environmental sustainability and public health protection.

Supervisor: William Akuffo

Project Description:

The project seeks to design and develop a low-cost, modular autonomous racing platform that can be used to advance both research and education in autonomous driving. Built on a small-scale race car, the system will integrate key components of embedded control, perception, and planning to mimic the functionality of real autonomous vehicles. This setup provides a practical and hands-on environment where theories in robotics, control systems, and artificial intelligence can be tested and applied.

Beyond its technical scope, the platform is intended to serve as an accessible testbed for experimenting with advanced real-time algorithms such as Model Predictive Control (MPC), sensor fusion, and reinforcement learning. By prioritizing affordability and modularity, the project lowers barriers to participation in high-performance autonomous driving research. Ultimately, it aims to empower students, researchers, and educators with a scalable tool for innovation, while encouraging collaboration and practical skill development in robotics and intelligent transportation systems.

Student: Samuel Muyambaze

Supervisor: Bright Tetteh

The scope of this project is to design, simulate, and prototype a wireless charging system for electric vehicles using inductive power transfer technology. It will involve the development of a transmitter and receiver coil system, integration of power electronics such as inverters and rectifiers, and implementation of a resonant inductive coupling mechanism to enable efficient energy transfer without physical connectors. A scaled-down prototype will be built to test and evaluate the system’s performance under different coil alignments and distances, focusing on parameters such as efficiency, transfer distance, and stability. The project will be limited to a laboratory-scale model and will not extend to full-scale commercial EV chargers or advanced grid integration.

Supervisor: Eleazar Archer

Project Description: This project focuses on designing and building a small-scale hydraulic press to expand hands-on engineering at Ashesi University. Right now, most student projects rely on 3D printing and basic manual methods, which limit durability and strength. The hydraulic press will solve this by allowing students to mould strong, precise parts such as brackets and enclosures, making it possible to move beyond prototypes and develop real, scalable products.

The press will generate at least ten tons of force and will be designed, fabricated, and tested to ensure safety, reliability, and accuracy. Along with the machine, the project will produce clear documentation, including a user manual, technical drawings, and design files, so it remains a lasting resource for future students.

In the end, this project is not only about delivering a functional tool but also about creating a learning experience that shows how engineering theory can be turned into practical solutions. It will strengthen innovation on campus and leave behind both a valuable machine and a roadmap for future projects.

Supervisor: Kofi Adu-Labi

Current outdoor air quality monitors often rely on grid or solar power, limiting their deployment flexibility. There is a critical need for a portable and rapidly deployable solution to perform short-term monitoring in remote or temporary locations. This project aims to design and prototype a compact, battery-powered IoT air quality device capable of measuring key pollutants and operating for a minimum of 24 hours on a single charge.

Supervisor: Dr Elena Rosca

Project Description
This project aims to design and develop a food scanner that can detect harmful bacteria on fresh fruits and vegetables using an integration of optical sensors, gas sensors, and deep learning models. Images will be captured, and details such as shape, size, and light absorption from the food surface will be processed through a trained deep learning model that can identify patterns linked to spoilage and bacterial contamination.

The model will be deployed on embedded hardware so that real-time predictions can be made without external processing. The device will present results on an LCD screen showing the type of bacteria detected and the percentage of food spoilage. In addition, the system will provide an optional UV light sterilisation feature that can reduce microbial load after detection.

This project will benefit food vendors, market women, households, and health inspectors by providing a rapid, reliable, and affordable tool for monitoring food safety. The proposed design will combine hardware and artificial intelligence to create a compact solution that advances public health and supports safe food handling practices.

Visual Description of Project: https://lucid.app/lucidspark/5fb9d339-68c8-43c2-bd0e-9d4d2b12d5d6/edit?viewport_loc=-1892%2C-488%2C4608%2C2085%2C0_0&invitationId=inv_5f6bc62d-a39b-4a10-9b61-210b0a8dff79

Supervisor: Elena Rosca

Project Description:

Asthma is a condition that causes chronic inflammation within the airways. This inflammation makes asthmatic patients susceptible to triggers such as smoke, pollen, exercise, or even cold air. When triggered, the airways experience a bronchospasm (they narrow), swell up, and fill up with mucus - this makes it hard to breathe resulting in coughs and wheezes. Without proper treatment, these attacks can be fatal.

The World Health Organisation projects that about 300 million people are suffering from asthma worldwide with approximately 250,000 asthma-related deaths annually. The last few decades have witnessed a rise in asthma, especially within industrialized nations. Further studies indicate that asthma is not a prevalent issue for high-income countries but is increasingly becoming widespread in developing countries, with most asthma-related deaths occurring in lower and low-income countries within Sub-Saharan Africa.

The system operates using a 50 W solar panel that charges a 12 V battery through a charge controller. Power from the battery is used to run an Arduino Uno microcontroller and a thermal imaging camera. At regular intervals, the Arduino activates the camera to scan the PV panel surface, capturing temperature variations. The thermal data is then processed to identify abnormal heating zones that indicate the presence of a hotspot.

Once a hotspot is detected, the system can trigger alerts through different means, such as a buzzer, LED indicator, or wireless communication module, to notify technicians or system operators. This proactive approach allows for timely cleaning, repair, or replacement of affected panels, thereby minimizing energy losses and extending the operational lifespan of the solar array.

By combining renewable energy power supply, embedded system control, and thermal imaging technology, PV Shield provides an affordable, autonomous, and sustainable solution for improving the reliability and safety of solar power systems.

Supervisor: Richard Awingot Akparibo

Project Description:

The characterization of photovoltaic (PV) modules is essential for monitoring performance, diagnosing faults, and improving overall energy yield. Traditional characterization methods often require external electronic loads or laboratory-based testing, which interrupt the module’s normal operation. This project proposes the design and implementation of an automatic electronic load system capable of varying the load applied to a PV panel in real-time. The system will electronically adjust the load profile, thereby enabling continuous measurement of current-voltage (I–V) characteristics without manual intervention.

The automatic load variation will be achieved using a controlled power electronic interface that emulates different resistance levels dynamically. By sweeping through a range of operating points, the system will extract key PV parameters such as short-circuit current, open-circuit voltage, maximum power point (MPP), and fill factor. The approach combines the precision of laboratory electronic loads with the adaptability of embedded digital control, making it possible to perform in-situ PV characterization under actual operating conditions.

The outcome of this project will be a prototype electronic load that is compact, programmable, and cost-effective, suitable for integration into solar monitoring platforms. The system has potential applications in renewable energy research, solar farm diagnostics, and educational laboratory setups.

Supervisor: Awingot Richard

Project Description:

Most irrigation and spraying systems are manual, rely on gasoline, and labor-intensive, leading to low yields and resource waste. This project develops a solar-powered automated hybrid system that performs both irrigation and spraying without manual effort, cutting reliance on gasoline-powered tools. It uses weather data to irrigate only when rain is not expected and automatically mixes chemicals based on type, ensuring precision, efficiency, and sustainability.

Student: Akua Oforiwaa Omane-Manu

Supervisor: Dr. Stephen Armah

Project Description:

This capstone project involves the design and fabrication of a semi-autonomous tank cleaning device for the oil industry. The aim is to eliminate human entry into confined spaces, which exposes workers to toxic gases, chemicals, and unsafe environments. The device will provide a safer and more affordable alternative to fully automated systems by combining an effective cleaning mechanism for oil residues and sludge with semi-autonomous controls. The outcome of this project is to improve worker safety while offering the industry a practical and cost-effective solution for tank maintenance.

Supervisor: Dr. Nathan Amankwah

Project Description:

This project develops a low-cost ($50) wearable system that continuously tracks vital signs heart rate, SpO₂, temperature, weight, and movement to detect early signs of heart failure decompensation, atrial fibrillation, or falls in elderly patients.

Unlike commercial smartwatches that depend on cloud processing, our device uses Edge AI on the ESP32 microcontroller to analyze data locally, enabling real-time alerts even without internet. It provides personalized insights, reduces false alarms, and can notify caregivers via SMS during emergencies.

By combining IoT, TinyML, and healthcare analytics, this innovation aims to make 24/7 cardiac monitoring accessible, private, and life-saving for older adults, especially in low-resource settings.

Supervisor: Richard Awingot Akparibo

Project Description:

Battery reliability is critical across off-grid solar systems, electric vehicles, and backup power installations. However, most battery health monitoring today relies on basic voltage and current metrics, which are insufficient for early-stage fault detection or accurate State-of-Health (SOH) estimation.

Electrochemical Impedance Spectroscopy (EIS) is a proven method for probing internal battery characteristics — including charge transfer resistance, electrolyte conductivity, and double-layer capacitance — which correlate strongly with SOH and degradation mechanisms. Unfortunately, EIS systems are typically lab-bound, expensive, and not designed for field use.

This project proposes a portable, microcontroller-based EIS diagnostic tool that provides interpretable battery health assessments in field conditions. The tool aims to bridge the gap between low-cost handheld testers and full laboratory EIS systems, making advanced diagnostics accessible to field technicians and researchers alike.

Supervisor: Dr. Stephen Armah

Supervisor: Dr. Elena Rosca

Supervisor: Dr. Elena Rosca, Lloyd Teta

Supervisor: Dr Elena Rosca

Project Title

Design of a Smart Solar-Powered IoT Sanitary Pad Dispenser for Schools and Communities in Ghana

Supervisor: Kofi Adu-Labi

Supervisor: Dr Elena Rosca

Student: Jada Abena Animwah Boateng

Supervisor: Sullaiman Alhassan

Student: Jada Abena Animwah Boateng

Supervisor: Sullaiman Alhassan

Supervisor: Master Bright Tetteh

This project seeks to design and implement a Smart Residential Load Management System capable of automatically scheduling and prioritizing household appliances based on Time-of-Use (TOU) tariffs, power availability, and user-defined preferences. The system integrates sensors, relays, and a microcontroller to monitor energy consumption, detect supply conditions, and intelligently control appliance operation in real time. Through a simple user interface, homeowners can view energy usage, adjust appliance priorities, and monitor when the system switches between grid power and a renewable energy source such as solar.

By incorporating TOU-based scheduling, load prioritization, and dual-source input management, the system provides a practical approach to reducing electricity costs and improving energy efficiency in Ghanaian households. During peak hours or limited supply, essential loads remain active while non-essential loads are deferred, ensuring stability and comfort. Overall, this project delivers a reliable demand-side management solution that promotes smart energy usage, reduces grid dependency, and supports the integration of renewable power in residential settings.

Supervisor: Dr Armah.

Supervisor: Dr. Stephen Kofi Armah.

Supervisor: Dr Stephen Kofi Armah

Supervisor: Dr. Stephen Kofi Armah

Supervisor: Prof Nathan Amanquah

LoRaWAN networks are privately owned but increasingly shared among different users and organizations. This raises critical challenges of trust and fair billing between network owners and users.

This project explores how blockchain technology can provide a decentralized, tamper-proof, and transparent framework for managing LoRaWAN networks. By integrating blockchain smart contracts, the system ensures that:

• Network usage is securely tracked.

• Owners can bill users fairly.

• Participants benefit from a trusted, decentralized infrastructure for IoT connectivity.

Student: Emmanuel Basakeng

Supervisor : Dr. Nathan Amankwah

Supervisor : Dr. Nathan Amankwah

Modern brain implants face a critical problem: wires must pierce through the skull to deliver power and extract data, causing infections and tethering patients to machines. In 2023, Purdue researchers solved this with Bi-Phasic Quasistatic Brain Communication (BP-QBC)—a method that uses gentle electrical signals through brain tissue to wirelessly power implants and transmit data simultaneously, achieving 100x better efficiency than radio waves, light, or ultrasound. My research adapts this breakthrough from brain implants to everyday wearable health devices. I'm building a system where a chest-worn hub sends power through your body tissue to a wrist sensor 50 centimeters away eliminating batteries entirely. The wrist device harvests energy directly from your body's natural conductivity while sending heart rate data back through the same tissue pathway. This solves the #1 complaint about health wearables (constant charging) by proving that the physics enabling wireless brain implants can also power devices on your skin. For elderly patients with heart disease or diabetes, this means continuous monitoring without ever worrying about dead batteries potentially preventing hospitalizations through gap-free health data. The project demonstrates how cutting-edge neurotechnology can translate into practical solutions that improve millions of lives.

Supervisor: Dr. Stephen Armah

Supervised by Acheampong Antwi Afari

Student: Emmanuel Bilson

Supervisor: Dr. Richard Awingot

Supervisor: Dr. Elena Rosca

Supervisor: Kofi Adu Labi