https://github.com/juliagarriga/arpet

AR.Drone 2.0 human tracking with Mobilenet-SSD and PID control
https://github.com/juliagarriga/arpet

drone mobilenet-ssd opencv ps-drone-library tracker tracking

Last synced: 6 months ago
JSON representation

AR.Drone 2.0 human tracking with Mobilenet-SSD and PID control

• Host: GitHub
• URL: https://github.com/juliagarriga/arpet
• Owner: juliagarriga
• License: mit
• Created: 2018-06-27T07:28:53.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2019-09-19T15:42:54.000Z (about 6 years ago)
• Last Synced: 2024-04-05T15:51:50.452Z (over 1 year ago)
• Topics: drone, mobilenet-ssd, opencv, ps-drone-library, tracker, tracking
• Language: Python
• Homepage:
• Size: 26 MB
• Stars: 14
• Watchers: 2
• Forks: 5
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# ARPET

Implementation of an onboard vision system autonomous tracker for an AR.Drone 2.0 using three PID controllers and the PS-Drone http://www.playsheep.de/drone/ library. The tracking algorithm is composed of an object detector implemented using a Caffe trained network of MobileNet-SSD, and a KCF object tracker using OpenCV v3. A Kalman filter is implemented to reduce the noise induced by the tracker. The code was tested indoors and the system yields acceptable results.

All code is written in Python 2.7. The dependencies are the PS-Drone library, OpenCV version > 3.0 and Numpy.

A visualization of the framework is shown in https://www.youtube.com/watch?v=ydnCSK7LFvU&t=2s.