Skip to content

Autonomous Landing of Quadrotor based on Computer Vision

Notifications You must be signed in to change notification settings

CatInTheRain/UAV_landing

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 

Repository files navigation

Autonomous Landing of Quadrotor based on Computer Vision

The purpose of this project is to have the drone perform a landing pad detection procedure and subsequent auto-landing under conditions such that ArUco fiducials cannot be detected.

Specifically, the objectives of the project are:

  • Landing platform detection for heights where ArUco fiducials cannot be identified.
  • Automatic landing from a generic initial position where the platform is detectable but the fiducials are not. The drone will have to arrive at a position such that it is able to detect the fiducials.
  • The drone will have to be able to find the platform again in case due to adverse phenomena the marker should disappear from the camera FOV. -he drone will have to perform a search maneuver for the platform in case an automatic landing is to be performed even though it has not yet seen the platform in the current scenario.

image

image

When centering on the marker in the horizontal plane, following control actions in velocity, the backhoe body rotates in roll and pitch. In the camera image, an apparent motion of the platform originates, and this generates unwanted oscillations in the references for centering. To overcome this phenomenon, the references to be sent to the controller, in order to generate the desired control velocities, must undergo compensation according to 𝝋 and 𝝑 angles.

image

Software-in-The-Loop Simulation running in Unity

image

K_1_5_sat_1_0_noARUCO.mp4

Real Experiment with Drone hardware

Media1.-.Trim.mp4

The complete report and code are available and can be requested in private.

About

Autonomous Landing of Quadrotor based on Computer Vision

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published