Open-UAV/openuav-playground

Cloud-enabled UAV swarm testbed. This repo has docker install files and a simulation sample with leader-follower behavior. Read more about the project at https://openuav.us

Project website: https://openuav.us

Here are the instructions for installing the simulation testbed on an Ubuntu 16.04 machine with an NVIDIA graphics card. Our test box is a Lenovo C730 with an NVIDIA GeForce RTX 2070 graphics card. Ensure that you have correct drivers installed by searching for and running the appropriate binary. For us, it was NVIDIA-Linux-x86_64-410.93.run found at https://www.nvidia.com/Download/index.aspx?lang=en-us .

• curl -sSL https://get.docker.com/ | sh

• sudo usermod -aG $USER docker

• sudo systemctl start docker

• sudo apt-get install docker-compose

• git clone https://github.com/Open-UAV/openuav-playground

• xhost +local:

• Do understand the implications of sharing xserver.

• curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list

• sudo apt-get update

• sudo apt-get install -y nvidia-docker

• pip install nvidia-docker-compose

• cd ~/openuav-playground/openuavapp

• nvidia-docker-compose build

This will take a while.

Once the build is complete, you will be able to run simulations and see it by running gzclient and rqt in the simulaiton container. Example:

• nvidia-docker run -dit --net=openuavapp_default --name=openuavapp_x${3:-date +%s} -v /tmp/.X11-unix:/tmp/.X11-unix -v /home/jdas/openuav-playground/samples/formation/:/simulation -e DISPLAY=:0 --entrypoint "/home/setup.sh" openuavapp_openuav

Assume the assigned simulation container name is openuav_x12345678 , you can verify yours by using docker ps

• docker exec -it openuav_x12345678 bash
• # source /usr/share/gazebo-7/setup.sh
• # gzclient

Gazebo client will pop up on your host.