Smart Controlled Unmanned Water Vehicle for Human Detection
Software as a Service
NoCategory
Thesis DesignClients
Techstack
Purpose
In a water emergency, sending a person to search is slow and can put another life at risk. A research team wanted an unmanned water vehicle that could spot people in the water on its own and report exactly where they are — buying rescuers time when every second counts.
Description
We built a camera-equipped water vehicle that runs live object detection on board to pick out a person in the water, splitting the video capture and analysis across parallel processes so it stays responsive on modest hardware. The moment someone is detected, it sends their location out by text message and email, while a web dashboard streams the annotated video and logs every detection for review. It is a demanding real-time computer-vision problem on an autonomous platform — the kind of build we take on and ship.
