Research direction

Research built around impact.

This page turns your profile into something scholarship committees and project evaluators understand immediately: what problems you work on, how you work, and why the work matters.

Core workflow
1

Problem selection: choose a real-world constraint such as mobility, cost, or limited access.

2

Data strategy: collect, annotate, and structure data so the model can actually learn useful behavior.

3

Model development: train and compare computer vision or ML approaches with practical metrics.

4

Deployment: optimize for edge devices like Raspberry Pi and keep the solution fast, offline, and stable.

Research themes

The site positions you as someone working across accessibility, edge AI, object detection, deployment engineering, and affordable robotics. That is much stronger than a vague “interested in AI” statement.

Assistive technology Edge inference Object detection Sensor fusion Low-cost design Dataset quality
Research angle Build systems that keep working without internet, without expensive hardware, and without sacrificing safety or usability.

What is being optimized

Latency, reliability, battery usage, affordability, user feedback clarity, and deployment simplicity.

Why the work matters

It is the difference between a demo and a device that can genuinely help people in the field.

Methods used

Python, PyTorch, YOLOv8, OpenCV, CAD, electronics prototyping, and iterative testing against real constraints.

Next direction

More rigorous benchmarking, better datasets, cleaner hardware integration, and sharper product-level presentation.