I am an Electrical Engineering Graduate and independent robotics researcher from Bangladesh. My primary research interest is around embedded AI applied to cost-effective robotics, especially robotic prostheses and exoskeletons. Additionally, I also focus on robot navigation and 3D reconstruction for semantic mapping. I co-developed TinyExo, a tinyML-based gait diagnosis unit that can reduce the processing unit cost of robotic prostheses and exoskeletons by up-to 80%. Furthermore, I also co-developed L-VITeX, a very lightweight vision-based heuristic system for visual robot navigation that can detect regions of interest with more than 90% accuracy with a Peak RAM Usage under 50 KB. This work has been featured by Hackter.io, the world's largest hardware developer community. As a testament to my devotion to robotics, I co-led my university's Mars rover team to the finals and 11th Global rank in International Rover Design Challenge 2023. Currently I am pursuing a Master's in EEE at the renowned BRAC university, Bangladesh where I research on Embedded AI for Robotics.
L-VITeX: Light-weight Visual Intuition for Terrain Exploration
Aug 2024- Sep 2024
L-VITeX aims to provide a hint of Regions of Interest (RoIs) without computationally expensive processing. By utilizing the Faster Objects, More Objects (FOMO) tinyML architecture, the system achieves high accuracy (>99%) in RoI detection while operating on minimal hardware resources (Peak RAM usage < 50 KB) with near real-time inference (<200 ms). The paper evaluates L-VITeX’s performance across various terrains, including mountainous areas, underwater shipwreck debris regions, and Martian rocky surfaces. Additionally, it demonstrates the system’s application in 3D mapping using a small mobile robot run by ESP32-Cam and Gaussian Splats (GS), showcasing its potential to enhance exploration efficiency and decision-making.