DTU-R3: Docker + ROS

DTU-R3 “Remote Reality Robot” Docker images based on ROS “Robot Operating System”.

Made for ARMv8 (Raspberry Pi 3) and AMD64 architectures. Tested respectively with Raspbian 9 Stretch, and Ubuntu 16.04 Xenial Xerus.

• Based on https://hub.docker.com/_/ros/ ros:kinetic-ros-base-xenial
  • http://wiki.ros.org/docker/Tutorials/Docker
• Images https://hub.docker.com/u/dtur3/
• Sources https://github.com/DTU-R3/Docker-ROS

Install Docker

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

Build DTU-R3 Docker images locally (optional)

Instead of fetching them from Docker Hub:

sudo ./docker-build.sh

To push the builds to Docker Hub:

sudo ./docker-push.sh
sudo ./docker-manifest.sh

Run ROS server and master

sudo docker run -dit --restart unless-stopped --log-opt max-size=10m \
	--network host --uts host -p 11311:11311 \
	--name ros_master ros:kinetic-ros-base-xenial \
	roscore

# See the list of topics:
sudo docker run -it --rm \
	--network host --uts host \
	ros:kinetic-ros-base-xenial \
	rostopic list

# Additional commands:
sudo docker ps -a
sudo docker logs -f ros_master
sudo docker stop ros_master
sudo docker rm ros_master
sudo docker restart ros_master
sudo docker exec -it ros_master /bin/bash
sudo docker run --rm weshigbee/manifest-tool inspect dtur3/r3-tutorials
# Cleaning
sudo docker images -a
sudo docker system prune
sudo docker system prune -a
# Updating
docker pull ros:kinetic-ros-base-xenial

Test basic ROS communication

See tutorials.

Run Parallax ArloBot

See ArloBot.

Run remote-control (keyboard, joystick)

See teleop.

Run waypoint navigation

See navigation.

Navigation requires a positioning system:

• See MQTT bridge to receive position information from Games on Track ultrasound indoor positioning.
• See Fiducials for 2D-code SLAM “simultaneous localization and mapping” from a camera.
  • See RaspiCam to expose a Raspberry Pi camera to ROS.

Visual programming

• See Node-RED.
  • See WS Bridge to expose a ROS to WebSocket bridge.