This week, I presented all my progress on the mobile application. Dr. Judhi identified some areas for improvement, particularly regarding the chatbot's recommendations, which were not as accurate due to a suboptimal prompt structure. Before the next meeting, I planned on refining the chatbot prompt to enhance its accuracy. I also showcased an animation and tutorial page, which Dr. Judhi found impressive but suggested adding an arrow indicator for rotation directions that were not clearly visible in the animation. Additionally, I demonstrated the successful storage of past shot data, which was now displayed on the profile page. Dr. Judhi recommended adding shot analysis text to this page as well, like in the dashboard for enhanced clarity. To further improve the project, I proposed a feature allowing users to compare their shots over time. Dr. Judhi found this idea highly valuable, and implementing this comparison feature became my next target before proceeding to user testing and evaluation next week.

During my meeting with Dr. Judhi this week, I informed him that the device had started encountering memory errors again, preventing shot data from being sent and stored in the Firebase real-time database. To resolve this issue, I explored different solutions and eventually implemented a circular buffering technique, which successfully restored the functionality without overflowing the buffer memory when the device was being used for a prolonged period of time. Due to the time spent troubleshooting this problem, my progress on the mobile application was somewhat limited. However, I was able to show Dr. Judhi the landing page of the application that I had worked on, along with the login and signup functionality of the application. Initially, I planned to develop the application using Unity, but after further research, I decided to switch to React Native using TypeScript. This choice was driven by React Native’s strong mobile application development focus, allowing cross-platform development for both iOS and Android without requiring a MacBook for iOS compatibility. Dr. Judhi appreciated my research-driven approach and reviewed my work so far. Additionally, I worked on a quiz page to collect additional information after taking a new shot, although it was not yet connected to Firebase. Dr. Judhi provided valuable UI feedback, which I will incorporate, and he instructed me to plan for the upcoming weeks, focusing on Firebase integration for seamless data exchange between hardware and software, along with developing the main pages of the app.

This week, I demonstrated to Dr. Judhi that I was successfully able to capture the shot data one second before taking the shot, and one second after the shot by implementing custom fluctuation calculations and setting thresholds. The device effectively measured the rapid acceleration and deceleration due to the force at the moment of striking the cue ball, which triggered the function to record and send the measured values to the Firebase Real-time database via a Wi-Fi connection. Dr. Judhi was particularly interested in testing the functionality himself and took a shot using a taped tennis ball due to the unavailability of a billiards table. Initially, the device data collection faced some issues, which I resolved the next day by fixing the memory mapping issue. Once the adjustments were made, Dr. Judhi took another shot, and the device functioned as intended. Additionally, the sensor was properly placed in the 3D printed mount, ensuring accurate shot data measurements. With this milestone achieved, Dr. Judhi advised me to now shift my focus towards the app development and plan for its integration with the hardware in the upcoming weeks.

This week’s session by Sir Roshan focused on UML diagrams and their role in software and system design. We explored how Use Case Diagrams help define the core functionalities of a system, identifying primary and secondary actors and their interactions. By mapping out these relationships, Use Case Diagrams serve as a visual blueprint, making it easier to communicate how a system works. Additionally, we discussed the importance of refining these diagrams iteratively to ensure they align with user needs and system goals. Moreover, we discussed about the other UML diagrams as well and how each of them can fit and convey information about different projects. For iCue, I developed a detailed Use Case Diagram, outlining how the ESP32 with MPU-6050, the mobile application, and the machine learning model interact to capture cue stick motion, analyze shots, and provide training recommendations to the user. This diagram plays a crucial role in ensuring a clear system architecture, guiding the next development phases. iCue Use Case Diagram I also designed a detailed Activity Diagram that illustrates the step-by-step flow of interactions between the ESP32 with MPU-6050, the mobile application, and the machine learning model. The diagram outlines how shot data is captured, processed, and analyzed, leading to real-time feedback and training recommendations for the user. This structured visualization helps in refining the system workflow, ensuring smooth integration between hardware and software components, and guiding the next stages of development. iCue Activity Diagram Tabtakir: A Celebration of Innovation & Entrepreneurship During our lab session, we attended Tabtakir, an event celebrating UAE Innovation Month. The event featured two incredibly insightful speaker sessions: 🎤 "Entrepreneurial Mindset and Resilience" by Dr. Ilham – A powerful session on how resilience, adaptability, and problem-solving are key to succeeding in entrepreneurship. 🎤 Yasmine’s Story with SoFreeOrganics – Yasmine shared how she built a successful business from the ground up without prior entrepreneurial experience, balancing it with being a full-time mother. Her journey was truly inspiring, showcasing how passion and determination can turn an idea into reality. These sessions deeply resonated with me, especially as I continue developing iCue. The entrepreneurial mindset is essential not just for starting a business but also for innovation in tech projects. Yasmine’s story reinforced the idea that with the right vision, dedication, and problem-solving approach, any project can evolve into something impactful—a philosophy I aim to apply as I work on iCue. Another highlight of Tabtakir was seeing the projects of my fellow juniors showcased at the event. As their Student Learning Assistant (SLA), I had the privilege of guiding them throughout their development stages, and it was amazing to witness their hard work being recognized. Seeing them present their innovations with confidence was truly rewarding. Project Progress: 3D Printing & System Integration This week, I made significant progress on iCue’s hardware development: Designed & 3D-printed the device mount using the RoboTechX lab’s 3D printer. Added a protective layer to prevent damage to the cue stick from the plastic mount. Successfully attached the mount to the cue stick, ensuring stability and precision for motion tracking. Explored an alternative screw-based design for the mount, which could allow for easier adjustments and upgrades in the future. With the physical framework now in place, my next steps will focus on: Capturing motion data only when a shot is detected, optimizing efficiency. Transmitting and storing all collected data in Firebase for further processing by the machine learning model. Reflection & Looking Ahead This week was very productive in terms of the project's technical advancements. Finalizing iCue’s hardware design, mapping out the system structure, and attending an event that celebrated innovation and resilience gave me fresh perspectives on my project’s future potential. The lessons from Tabrtakir reinforced the importance of resilience, adaptability, and continuous learning, all of which are key to developing iCue into a functional system. Next week, my focus will be on optimizing data collection and transmission, refining how iCue detects shots and stores data in Firebase. With both the physical and software architecture progressing steadily, I’m excited for the next phase of development!

I showed my gripper design to Dr. Judhi this week. After examining it, he graciously allowed me to utilise the 3D printer in the RoboTechX lab to print my device mount. I also showed my Use-case diagram for feedback which he approved. After the print was ready, I visited him again to show him the 3D-printed mount. He suggested adding a protective layer on the cue stick, such as paper, to keep the plastic from scraping the surface of the cue stick. I then proceeded to make this adjustment and successfully fastened the mount to the cue stick before presenting it to him again. He was pleased with the results and suggested an alternative design in which the device holder could be attached to the mount using screws if time permits later after the successful completion of the deliverables. Moving forward, Dr. Judhi urged me to focus on two crucial areas next week: Data is collected only during the spike (when a shot is detected). Sending all captured data to Firebase for further processing later. These will be my next steps as I continue to refine the project next week.

This week’s session led by Sir Roshan focused on prototyping techniques and their role in project development. We explored different approaches, ranging from low-fidelity (Lo-Fi) prototypes, such as pen-paper sketches and cardboard models, to high-fidelity (Hi-Fi) prototypes, which include digital mockups, coded prototypes, 3D models, and simulations. One of the key takeaways was that the choice of prototyping technique depends on multiple factors, including time, resources, testability, and realism. The session also introduced various prototyping tools, such as Figma, TinkerCAD, and FreeCAD, which can be used for different types of projects, whether they involve mobile applications, websites, augmented reality, or electronics. For my project, iCue, a combination of digital modelling and physical 3D printing is the most suitable approach. The device’s mount needs precise design and structural integrity, making 3D modelling and high-fidelity prototyping essential to ensure that the device is securely positioned and functions correctly. A Special Milestone: Celebrating 10 Years of MDX Computing Society & SLEJ This week wasn’t just about technical progress, it was also a memorable moment for the MDX Computing Society and the university’s tech community. During our lab hours, the class attended the MDX Computing Society and SLEJ’s 10-year anniversary celebration, an event I had the honour of leading and hosting. The celebration reflected on the milestones and achievements of MCS and SLEJ over the years and included an award ceremony for the 2024/2025 SLEJ winners. One of the biggest announcements at the event was the formation of the Middlesex University Dubai BCS Student Chapter, which I have the great honour of being the Founder and Chair of. This marks a significant step in connecting students with professional development opportunities, industry networking, and global BCS initiatives, something that projects like iCue can greatly benefit from. Project Progress: Optimizing Data Collection This week, I made significant progress on the hardware side of iCue. After troubleshooting and refining my implementation, I successfully: Fixed the MPU-6050 sensor issues, ensuring it now captures and displays all six values correctly. Established stable Bluetooth connectivity with the ESP32 and verified that sensor data is transmitted when powered via a LiPo battery connected to a TP4056 module. Optimized data collection to ensure values are recorded only when a spike is detected, indicating that a player is taking a shot. Refined the shot analysis approach, focusing on measuring how straight a player's stroke is and comparing it to a professional shot for personalized recommendations. With these improvements, my next task is to design and 3D-print a mount that will securely attach the sensors at the end of the cue stick, ensuring precise motion capture and stability. Reflection & Looking Ahead This week was the ideal combination of technical progress, strategic learning, and leadership milestones. Refining the hardware setup for iCue was a significant accomplishment, paving the path for accurate motion tracking and meaningful gameplay analysis. Understanding various prototype techniques also assisted me in determining how to construct the mount efficiently while keeping testing and usability in mind. Moreover, my plan is to utilize one of the prototyping tools discussed in class for modelling my mount next week. Beyond project development, leading the MDX Computing Society & SLEJ anniversary event and launching the Middlesex University BCS Student Chapter was a proud moment. It reinforced my passion for tech leadership, entrepreneurship and community-building, and I’m excited about the impact these initiatives will have on students and their ideas in the future. For next week, my focus will be on finalizing the 3D model for the device mount and beginning testing with the optimized data collection process. With steady progress, iCue is shaping up to be an exciting and impactful project!

This week, unlike the previous two, did not involve an ordinary classroom session. Instead, we were given the amazing opportunity to visit the AI for Everything exhibition at Expo City Dubai. The event showcased the most recent advances in artificial intelligence, providing important insights into how AI is influencing businesses and tackling real-world challenges. It was an inspirational event that broadened our understanding of AI's possibilities and allowed us to network with industry experts. Many of the discussions at the event focused on how AI is transforming industries, and this resonated with me as I continued work on iCue. Seeing real-world AI applications reinforced the importance of data-driven insights in sports technology. It also sparked new ideas on how machine learning could be further optimized for analyzing player movements and improving training recommendations. One of the event highlights was the Middlesex Innovation Hub's (MIH) own booth. Along with my group members Sana, Ninas, and Arsal, I had the opportunity to present our GuardianEar project from our second-year Digital Signal Processing module at the MIH booth. GuardianEar is a smart device that detects spy microphones and determines how far they are from the device. Presenting our project in front of professionals and fellow students was a great experience that allowed us to demonstrate how GuardianEar uses signal processing techniques to make a real difference. The feedback we received was very positive and constructive. Hoping to take the project further with all the feedback received. The feedback also highlighted how AI models must be both accurate and user-friendly, a key takeaway I will apply while developing iCue’s recommendation system. The event concluded on a positive note, with our group finishing as 1st runner-ups in Unipreneur's Youth AI Builder Contest! Winning this accolade for GuardianEar was a proud occasion that highlighted the significance of our efforts. On the project front, I've begun work on the final formal proposal, opting to use LaTeX for a more polished and professional appearance. I revisited the LaTeX tutorials and started refining my document because I hadn't used it in a long time. I intend to submit the proposal early next week. In addition, I researched the best approach to put the components on the cue stick. Following some research and meeting with Dr, Judhi, I realised that attaching the components to the end of the stick is the best method. As the majority of the shot force is concentrated at the top of the stick, this method increases stability and reduces the risk of damage. Moreover, I will be able to get more accurate readings from the sensor due to better angle positioning of the sensor. Reflection This week was a perfect combination of exposure and progress. Presenting GuardianEar at a large-scale event and gaining recognition provided a significant boost in my confidence. It reminded me how much I appreciate working on projects that address real-world problems. The experiences from the event will definitely help motivate me as I continue my work on iCue. Beyond gaining recognition for GuardianEar, the AI for Everything event provided valuable insights that I can apply to iCue. Seeing how AI is used for performance tracking and real-time analytics inspired me to refine how iCue processes motion data and delivers recommendations. This experience reinforced my belief in leveraging AI to enhance skill development, and I’m excited to implement these learnings in my project. For iCue, I made substantial progress towards completing the formal project proposal. Revisiting LaTeX allowed me to improve my documentation abilities and arrange the proposal more professionally. After investigating the best mounting position for the hardware, I concluded that positioning the components at the end of the cue stick would result in more stability and data accuracy. This discovery is critical for the design phase, and I'll integrate it into my next steps. Overall, it was an eventful and productive week that laid the foundation for work ahead.

This week's session led by Sir Roshan concentrated on software development processes and the various frameworks employed during the project development life cycle. We looked at numerous development paradigms, including waterfall, prototyping, spiral, unified, prescriptive, and agile. We also discussed common development frameworks such as Scrum, XP, Kanban, and DevOps. A key takeaway from this week's session was the prototype evolutionary process, which emphasizes iterative improvements based on testing and feedback of the project. Given the nature of my project, Agile development appears to be the most appropriate method. Agile enables ongoing refining through incremental updates, making it suitable for iCue, in which the hardware, software, and machine learning models will be built concurrently and improved depending on user feedback. However, some components of the project, such as hardware assembly, require a more systematic approach, making Waterfall development process applicable in those areas. Project Progress: Assembling & Testing the iCue Device And the project work officially begins! This week, I focused on assembling and ensuring the iCue device was functioning properly before proceeding to data collection and analysis. I successfully set up the MPU-6050 gyro and acceleration sensor, but encountered an issue where it was only recording five values instead of six and failing to print error-handling messages in the serial monitor. Additionally, I had trouble uploading modified code to the ESP32, which stalled further development. To move forward, I met up with Dr. Judhi for guidance who suggested to start troubleshooting these issues, checking hardware connections, debugging the serial monitor output, and reviewing possible driver installation problems. These tasks are crucial for ensuring reliable data collection, as the sensor’s accuracy directly impacts iCue’s motion analysis and recommendation system. Along with setting up the hardware, I also managed to complete my formal project proposal using LaTeX. At the end of this blog, I have attached my proposal, which outlines iCue’s objectives, technical approach, and development plan in greater detail. Reflection & Next Steps This week was quite productive as it combined technical problem-solving with theoretical learning about software development processes. Understanding Agile and Waterfall methodologies helped me better plan my project, balancing iterative software developments with a systematic approach to hardware development. On the project front, while challenges arose in sensor communication and ESP32 programming, they provided valuable insights into debugging hardware-software integration issues. Moving forward, my focus will be on resolving the sensor’s data capture problem, ensuring all six values are recorded correctly, and successfully uploading modified code to the ESP32. Once this is done, I will proceed with testing data transmission and exploring the best ways to store and process the collected motion data. Formal Proposal (Also in Document Uploads Section)

During this week's supervisor meeting, I showed my blog to Dr. Judhi and informed him that I had uploaded all of the required proposals and ethics documents. Following that, I showed the final assembled device functioning. We then discussed the project's first phase, and I displayed the MPU-6050 sensor in action. Initially, a few errors stopped the sensor from recording all six value measurements. It recorded five measurements instead and did not print the error handling strings in the serial monitor. Additionally, I had problems uploading modified codes to the ESP32. Dr. Judhi was kind enough to review the issues and offered helpful advice on how to fix them. He suggested that the port did not have the right driver installed to enable communication. He then tasked me with figuring out the solutions to these issues and solving them before meeting him again next week.

This week marked the beginning of our final-year project journey with an introductory session led by Dr. Fehmida and Mr. Roshan. They gave us an in-depth overview of the module, including timelines, expectations, and deliverables for the following months. A key part of the lecture was a discussion of several research methodologies, with an emphasis on how each methodology can be used in our projects. The Build, Model, and Experimental methodologies stood out to me as the most applicable to my project. The Build methodology focusses on the design and development of my device's hardware and software. The Model methodology will be applied during the development of algorithms for classifying skill levels and providing feedback, whereas the Experimental methodology will be required for testing the device with participants and determining its accuracy. Mr. Roshan also introduced us to the First-Cut Proposal form, which allows us to define our project's goal, objectives, and scope. He discussed how this document will help us refine our project ideas and successfully communicate them to our supervisors. Following the workshop, we were instructed to contact our supervisors for an introductory meeting to discuss our plans and ideas. We were also given the task of completing the First-Cut Proposal form and starting with our blogging website. The Idea For my final-year project, I decided to work on iCue, a Smart Billiards Gameplay Analyser and Recommender. The project aims to create a smart device that attaches to a billiards cue stick to track movement and collect gameplay data. This device, when combined with a mobile app, will analyse the data and provide real-time feedback on shot accuracy, alignment, and consistency. It will also classify users by skill level (beginning, intermediate, or advanced) and provide personalised advice to assist them improve their gameplay. The Inspiration Behind iCue As someone who grew up playing all kinds of sports, I have always been fascinated by the intersection of technology and sports, particularly how smart systems can enhance training and performance analysis. Billiards is a game of precision, yet most players lack access to advanced training tools that provide meaningful feedback on their shots. Existing solutions are often expensive or focus on isolated aspects such as shot alignment without giving a comprehensive picture of a player’s skill level. That’s what inspired me to develop iCue. With iCue, I hope to offer an affordable and user-friendly system that bridges this gap and helps players improve their skills using technology-driven insights. The project will include hardware and software integration. Sensors will be used to track cue stick movements and provide data on shot alignment and cue stability. This data will be transmitted to a smartphone application, which will analyse it in real-time and provide feedback and suggestions for improvement. The plan is to build and prototype the device, develop the accompanying application, and test the system with users of varying skill levels to ensure accuracy and usability. Approach: Phase 1:  Research and design the hardware device and select appropriate sensors for motion tracking. Phase 2:  Develop the mobile application with an intuitive interface for data analysis and feedback. Phase 3: Develop the model to make predictions and recommendations. Phase 4: Integrate the hardware and software. Phase 5:  Conduct user trials to test the device and gather feedback for improvements. Methodology: After evaluating different approaches, I found that the Build, Model, and Experimental methodologies were the best fit for iCue. Build : Focuses on designing and developing the hardware and software components. Model : Applies to creating algorithms that classify skill levels and provide real-time feedback. Experimental : Involves user testing to evaluate the accuracy and effectiveness of the device. Reflection After spending a lot of time brainstorming and discussing with my supervisor, I finally settled on a project idea this week. Initially, I had various ideas, but after extensive research and careful consideration, I chose iCue because of its potential to make a significant difference in the billiards industry. I've also worked on my First-Cut Proposal attached below, which has helped me define the project's goals and strategy. Moving ahead, I intend to focus on the hardware prototype and finalise the app design. This is just the beginning, and I’m excited to see how iCue can transform the way players train and improve their gameplay! First-Cut Proposal (Also in Document Uploads Section)

In this week's meeting, I showed my progress since the last weeks meeting. I successfully rectified the issues I faced last week, allowing the MPU-6050 sensor to accurately measure and show all of the six values. I was also able to establish Bluetooth connectivity with the ESP32 and obtain the sensor values when the device was powered by a LiPo battery connected to a TP4056 module for charging the battery. Given the huge number of values collected, Dr. Judhi proposed optimising data collecting by recording values only when a spike is detected, suggesting that a player is taking a shot. To simplify the process, he suggested assessing how straight a player's stroke is and comparing it to that of a professional. This comparison would then be utilised to create recommendations for the player. Dr. Judhi was also very kind to provide his cue stick this week, which I can utilise to prototype my device. Based on the feedback this week, my objective for the following week is to implement the suggested optimisations and to 3D-print the model of the mount to securely attach the sensors at the end of the cue stick for precise cue stick motion capture.

In the meeting this week with Dr. Judhi Prasetyo, I presented my initial design idea for securing the components onto the cue stick. My original plan was to attach the components on top of the stick, but Dr. Judhi recommended a more practical solution. He suggested creating a 3D-printed mount that could be fitted to the end of the cue stick instead. This method would improve stability and durability because the vast majority of the shot power is concentrated at the top of the stick which would result in the device flying off the stick. Moreover, I will be able to get more accurate readings from the sensor due to better angle positioning of the components. After getting this insightful feedback, I'll now refine my design before beginning to create and print the 3D-printed mount.