Overview
Gagamba explores how a drone could inspect and help maintain tall buildings by seeing and tracking what’s in front of it in real time.
The engineering challenge
Autonomous systems need to interpret visual inputs and produce useful detections fast enough to support downstream decisions in a physical environment.
What I built
- Developed real-time object detection and tracking workflows using Python, OpenCV, and YOLO.
- Explored cloud-based model development and training workflows using AWS SageMaker.
- Investigated how computer vision could support autonomous inspection and navigation.
- Explored reinforcement learning as a potential approach to decision-making and control.
What this demonstrates
Real-time processing, applied machine learning, cloud ML workflows, and the intersection of software and physical systems.