HomeProjectsExperienceBlogsResume
 
  

OpenQuad

July 15, 2020

It can be used for implementing various deep learning and computer vision algorithms such as person tracking, Gesture recognition, Optical flow stabilization, Human Pose estimation, obstacle avoidance, and depth estimation using monocular vision. The drone uses a pixhawk flight controller with RaspberryPi as a single board computer. DJI flame wheel is used for the quadcopter structure with custom mountings for safety measures. Serial communication is used to communicate between Pixhawk and RaspberryPi. RaspberryPi runs a ROS node that communicates with the ROS node running on the host PC to transfer videos over Wi-Fi. To make the project open-source, easy to develop, and reproducible, simulation environment setup has been dockerized using docker container. Gazebo is used for simulation.

Image of construction and placement of componentsAnother Image of construction and placement of components

Controlled flight of the Quadcopter


Link to GitHub

Link to Docker Container

© Sarthak Narayan | 2022