Remote Nav
Authors: Suzanne Dazo, Alex Hubers
What it's for: Remote nav is the remote navigation interface for navigating a given robot with privacy filters.
In order to run, it needs ROS groovy, python 2.7, QT4, the custom_nav package (located in this repo), and the rviz library You can install the rviz plugin package by using:
sudo apt-get install ros-groovy-visualization-tutorials
On the robot of choice (known to work with both the Turtlebot and PR2) Install the custom_nav package beforehand
On the PR2:
robot claim
robot start
roslaunch custom_nav pr2_amcl.launch map_file:=/path/to/map.yaml
On the turtlebot:
roslaunch turtlebot_bringup minimal.launch
roslaunch custom_nav amcl_demo.launch map_file:=/path/to/map.yaml
On the computer:
ROS_MASTER_URI must be set to the robot for everything to be viewable. If you cannot see the robot in rviz, the UI cannot see it, either.
Make sure the workspace is sourced in groovy
source /opt/ros/groovy/setup.bash
#Navigate to the root of the workspace folder and run:
source devel/setup.bash
#Then run the correct file depending on the robot
roslaunch remote_nav pr2UI.launch map_file:=/path/to/pretty/map.yaml
roslaunch remote_nav turtleUI.launch map_file:=/path/to/pretty/map.yaml
#Where map_file points to a a map.yaml you wish to display.
#It does not have to be the map used in AMCL.As a note, known bugs and issues should probably be handled before the improvements, but this is the direction we eventually want to go with our interface and the Privacy Interfaces project as a whole.
- Make it so that you can click on the video feed of the robot, and the robot will look (or turn if Turtlebot) in the direction of the click. This requires getting click coordinates over the rviz frames in the UI
- Make transparent buttons that can lay over the interface instead of being placed below
- Convert the hard-coded interface into a QtDesigner .ui file
- Make it more open-ended so the robot can handle navigation around a room instead of along a specified path
- Be able to read in the output from the privacy_zones package and use that to limit movement to private areas
- Integrate with openCV instead of rviz in order to display video that has already been filtered for privacy