Bryan Kikuta
Back to Projects

Indoor Drone Localization System

May 2023 - August 2023

ROS
Python
Nvidia Jetson Nano
Pixhawk
Computer Vision
OpenCV
ArUco Markers
software

I developed an indoor drone localization system for GNSS-denied environments at 15% cost of traditional systems.

Key features:

  • Utilized a camera-vision based approach implemented with ROS on an Nvidia Jetson Nano computer
  • Integrated with a Pixhawk flight controller
  • Achieved accuracy of 30 cm on a multi-rotor drone indoors
  • Localized the drone globally by determining the positions of ArUco markers with OpenCV
  • Performed coordinate system transformations to determine the global position of the drone for autonomous flight
View on GitHub
Indoor Drone Localization System
Indoor Drone Localization System - Image 1
Indoor Drone Localization System - Image 2
Indoor Drone Localization System - Image 3
Indoor Drone Localization System - Image 4