Setting up RGBDSLAM with ROS in the real world

From Lofaro Lab Wiki
Revision as of 22:46, 9 February 2015 by Rtindel2 (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

1.) Plug the Kinect Sensor into the computer.


2.) Also plug the Pioneer 3dx into the computer.


3.) Also plug the joystick into the computer.


4.) Open a terminal and type:

  roslaunch rgbdslam odometry.launch

'''This will do the following:'''

spin up the openni2 node ​ spin up the rgbdslam node

spin up the octomap server

opens rviz with all topics subscribed correctly


5.) Open another terminal and type:

  sudo chmod 777 /dev/ttyUSB0
 
  rosrun rosaria RosAria port:=/dev/ttyUSB0

'''This will do the following:'''

Make the connection to the robot that will allow the control of the motors


6.) Open another terminal and type:

  sudo chmod 777 /dev/input/js0
  
  roslaunch joy_control pioneer_joy.launch
  

'''This will do the following:'''

This is a controller that will allow the user to control the motors using a joystick.


7.) We are now ready to start SLAMMing. Put your focus on rgbdslam and press the spacebar

'''This will do the following:'''

The mapping process will begin


8.) Use the joystick to begin moving the robot VERY SLOWLY.

'''This will do the following:'''

The map will begin to grow. You will see Landmarks on the bottom left of the RGBDSLAM window and they will be showing the Visual Odometry.


9.) Press the Add button in RVIZ and go to the "by topic" tab. Scroll down and choose the "/project_map" topic.

'''This will do the following:'''

RViz is now ready to visualize the map from the octomap server.


10.) In the RGBDSLAM window press CTRL+m

'''This will do the following:'''

The pointclouds will be sent to the octomap server. Rviz will begin to populate a 2D map.


11.) When you are finished press the Spacebar in RGBDSLAM.

'''This will do the following:'''

RGBDSLAM will stop processing


12.) To save the created map, open a terminal and type:

  rosrun map_server map_saver -f map map:=projected_map

'''This will do the following:'''

a map.pgm file and a map.yaml file will be save to file