Hallo - a hand controlled Tello

DJI Tello controller using hand gestures python app

Docker installation in 3 steps:

Tested on ubuntu 19.10

step 1

git clone github/GalBrandwine/hallopy.git
cd HalloPy

step 2

docker build --network=host  --tag hallopy:1.3 .

(make sure you have docker installed.)

step 3

(Make sure you're connected to the TELLO wifi)

xhost +  && docker run --rm -it --net=host --ipc=host -e DISPLAY=$DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix --privileged -v /dev/video0:/dev/video0 --name hallopy hallopy:1.3

For flags explanation: running docker with gui

Non docker users:

Environment:

Dependency libraries installation:

Run

From directory /HalloPy/:

python ./hallopy/hallo.py

Controller Usage:

  1. Make sure you have all dependency libraries. for great openCV installation tutorial refer to: https://www.pyimagesearch.com/2016/10/24/ubuntu-16-04-how-to-install-opencv/
  2. Turn on Tello drone and connect to it's wifi
    1. there's no setup.py yet, so in order to run this project, open the project in an IDE and run: hallo.py

  3. Application usage:

    press 'b' - to detect palm ( depend on the environment lights, a Thresh Hold slider tuning may be needed)
center the detected palm at the center of the detection_frame FOR 5 seconds 
pressing b again will reset calibration.

    After center-circle become GREEN (meaning we are now calibrated):

    press 't' - to take off ( a 3 second hold-up, until drone is in the air)
  after landing - if program is calibrated, press t again to take-off

press 'c' - to toggle drone's control between key-board and detected-palm
  toggling back to keyboard will force drone to hover in the same place

press 'l' - to land ( at any time)

press 'x' - to adjust background' threshold - for thicker palm recognition
press 'z' - to adjust background' threshold - for thinner palm recognition

press 'esc' to exit program ( only after drone landed)
  4. Video explaining hands movements for controlling the drone can be found here, and here

Tanks to: