/apple_picking_robot

Apple picking robot University of Latvia and Accenture co-project

Primary LanguageJupyter Notebook

Apple picking robot

Creating autonomous robot for given size apples pitching in the controlled indoor environment to gain and share the knowledge between industry and academy.

Project structure

  • data/ used throughout project. Usually we ignore data stored there from .git perspective.
  • docs/ project documentations. All topics we want to be documented. See Docs table of content.
  • Scientific papers/ and reviews. Usually you give and share papers using url but there are some papers which we may want to store in repository.
  • presentations/ by team members.
  • ros/ (Robot Operating System) used for inference with Virtual Environments or Robots
  • experiments/ consists of different experiments such as training neural network, preparing datasets etc.

Run

  1. Install Docker on your machine if you do not already have it. See more information under section #first time run

  2. Pull latest Docker image and run it from terminal:

    docker pull naurislv/apple_picking_robot
docker run -it -p 6080:80 naurislv/apple_picking_robot

  3. Open browser: http://127.0.0.1:6080 and open terminal inside

  4. Running project.

    • First navigate to ros directory: cd /home/apple_picking_robot/ros

    • Run ./run_apple_picker.sh --help to see all possible input arguments:

      Apple Picker

[options] application [arguments]

options:

-h,  --help                        show brief help
-rv, --rviz=bool                   if true then run ROS rviz GUI
-r,  --rqt=bool                    if true then run ROS rqt_image_view
-c,  --camera=bool                 if true then physical camera connected and no virtual env will be launched
-su, --scene_understanding=bool    if true then scene_understanding will be launched
-kb, --keyboard=bool               if true then run keyboard package for controlig Gazebo
-w,  --world=WORLD                 choose turtlebot world. empty | original | dbaby | None
-gg, --gazebo_gui=bool             if true then run gazebo GUI
-yt, --youtube=LINK                download youtube (LINK) video and copy it experiment directory

Example: ./run_apple_picker.sh -r true -kb true -w dbaby
Example: ./run_apple_picker.sh --rqt=true --keyboard=true --world=dbaby

    • Run script with arguments e.g.: ./run_apple_picker.sh -r true -kb true -w dbaby

Docs

  1. Robot (real and virtual environment) software and hardware architectures and related paper reviews.
  2. Intro to ROS
  3. How to use Docker for this project.
  4. UnrealEngine4
  5. How to setup Gazebo

The Team

In Progress: Create table of all team members

Image Role Name Location LinkedIn email
Nauris Dorbe ML, Autonomy Nauris Dorbe Latvia, Riga Nauris naurisdorbe@gmail.com

Our story

NIPS (Neural Information Processing Systems) conference 2017

We've been there with dBaby