My current research is in the interfacing of computer vision techniques with an existing robot prototype; FAll DEtection Robot (FADER) to carry out detection, distance estimation, navigation, tracking, and fall detection in that order.
The aim is to have such mobile robots in the homes of elderly people who live alone and to have it detect when they fall so they can get help fast. We are currently using both the Raspberry Pi 3 for control and the Rapsberry Pi Camera for computer vision (specifically Deep Learning Single Shot Detector(SSD) Framework with the MobileNets Base Model).
The advantages of our approach is that it is
- Non-participatory for detection and tracking
As at December 2018 (Fall 2018), I have achieved detection, distance estimation and navigation in 1D space(see video below). Next is 2D navigation and tracking.
FADER without the Pi Camera SSD Object Detection Running on the Pi Camera