Localization through RGB-D Camera for the Quad-rotor

Vidit Datt Prabhakar, Vijay Kumar Pandey, Anuj Kumar Sharma, Tushar Sandhan

Abstract: This paper presents a thorough investigation into the implementation of a localization system using a RGB-D camera that prioritizes computational efficiency. The study focuses on utilizing the Oak D Pro depth camera, which includes an in-built Inertial Measurement Unit. Visual Odometry is used to estimate the camera pose which is lighter in computation than other algorithms like Simultaneous Localization and Mapping. The IMU is used to obtain the camera pose due to the presence of drift during the camera’s fast motion. The readings are fused together to get the correct camera pose by employing the Extended Kalman Filter for sensor fusion, achieving seamless integration of VO and IMU. This research explores the adaptation of Visual Inertial Odometry system for an autonomous Unmanned Aerial Vehicles applications, specifically for a quad-rotor/drone.

External IDs:dblp:conf/icccnt/PrabhakarPSS24