Abstract:
The objective is to provide an effective solution for 3d mapping in GPS incapable
mission scenarios. This concept makes use of orb-slam (Systematic Localization &
Mapping) for robust object tracking and path planning. The monocular camera on the DJI
Tello drone enables a video transmission to a ROS machine for processing. We aim to use
visual-slam techniques to obtain a local map of the environment and achieve greater
accuracy along turns. With the help of computer vision techniques, localization is
performed on an inexpensive consumer quadcopter. This work serves as a model to achieve
Slam implementation across different drones. Using orb-slam vocabulary with a centralised
collaborative Slam enables the use of multiple drones. Robotic collaboration promises
increased efficiency of missions with great potential in applications, such as
search‐and‐rescue and Military operations.
