> INIT_VISION_MODULE...
> LOAD_KINEMATICS...
> SYSTEM: READY.
LON: 02°13' W
LAT: 53°27' N
MSc Robotics Student | Future Robotics Engineer
I am a passionate robotics enthusiast currently pursuing my MSc in Robotics at University of Manchester. With a background in Computer Science and Electric Engineering, I am dedicated to developing innovative robotic solutions that can make a positive impact on society.
My research interests lie in visual recognition for embodied systems: improving robustness of detection and recognition under noisy/low‑resource conditions, lightweight model design (pruning, quantisation, knowledge distillation) for edge deployment, and temporal/multi‑modal fusion to stabilise tracking in dynamic scenes. I’m also keen on tools that close the loop between algorithms and users — e.g., web‑based visualisation, synchronized annotation and human‑in‑the‑loop labelling to accelerate dataset curation and model validation.
To become a leading robotics engineer specializing in autonomous systems, contributing to cutting-edge research while developing practical solutions for industrial applications.
An autonomous mobile robot system integrating a Leo Rover to navigate through random environments, detect coloured objects, and pick-and-place them into corresponding target bins.
My role focused on the control and system integration of the robotic arm. I implemented the pick-and-place state machine workflow, and managed TF2 coordinate transformations between the vision system and the manipulator.
A robust feedback control system developed for the DJI Tello drone (simulated and real-world) to achieve highly precise waypoint tracking and orientation control.
Implemented a custom Disturbance Observer (DOB) combined with an adaptive PID controller to dynamically compensate for environmental disturbances (e.g., wind). Includes interactive 2D and 3D visualization tools for flight trajectory analysis.
A grid-based A* path planning algorithm implementation for autonomous mobile robots to find optimal collision-free routes.
Implemented the core A* algorithm in pure Python using a 4-connected motion model and Euclidean heuristic, integrating it with a GUI for visualization.
A comparative study of Deep Learning (PyTorch ResNet18) and traditional Computer Vision (BoW with SIFT/SVM) for image classification.
Evaluated the trade-offs between model accuracy, computational efficiency, and interpretability.
An immersive Web VR platform for multi-user collaboration and borehole data visualization.
Built by A-Frame, WebSockets, and AWS, featuring time-synced annotations and dual-video comparison.
A low-power voice authentication access control system for smart homes using Sony Spresense and Raspberry Pi.
My role involved implementing real-time audio processing (MFCC & GMM), optimized hardware tasks, and designed interactive LED feedback.
An academic presentation exploring the role and applications of robotics technology in achieving the United Nations Sustainable Development Goals (SDGs).
An experimental study analyzing the capabilities and limitations of a single 360-degree 2D RPLiDAR in detecting various spatial obstacles in realistic environments.
I'm always open to discussing robotics, research opportunities, or potential collaborations.