Please see our website here!
Put an AR tag on top of turtlebot and have the y direction of the AR point in the forward direction of the turtlebot.
Remember to set
export ROS_MASTER_URI=http://[TurtleID].local:11311
ssh turtlebot@[TurtleId].local password is EE106A19
then run
roslaunch turtlebot_bringup minimal.launch --screen
If you want to check things are working you can run teleop with:
roslaunch turtlebot_teleop keyboard_teleop.launch --screen
Grab a big tripod instead of the small one that comes with the realsense.
Make sure you have the realsense plugged into a USB 3 port through a USB 3 cable.
To run the perception module run the following command:
roslaunch realsense2_camera rs_camera.launch align_depth:=true tf_prefix:=realsense
roslaunch lab4_cam ar_track.launch output_frame:=realsense_color_frame
rosrun perception main.py --verbosity 0
You will need to have ar_maker_13 in the field of view of the camera. It uses this ar tag as the world frame and publishes the state estimate of the ball in this frame.
If you want to use a different ar tag use the --ar_tag_number argument:
rosrun perception main.py --ar_tag_number {ar tag number}
By default perception displays the RGB image, depth image, and mask and prints the
location of the ball. To remove this for improved speed use the --verbosity
argument see the constructor of RealsensePerceptionProcess for more info.
To view the ball location in rviz run
roslaunch turtlebot_rviz_launchers view_navigation.launch then
open the turtlebot_soccer.rviz configuration.
To check the speed of everything run
rostopic hz /camera/color/image_raw /camera/aligned_depth_to_color/image_raw /state_estimate /predicted_path /intersection_point
they should all be around 30 hz. Sometimes the camera will overheat and depth and/or RGB images will drop to 15 Hz. Unplug the camera and cool it down and then it should be back at 30 Hz.
rosrun prediction main.py
rosrun planning main.py --turtlebot_frame {turtlebot_frame}
where {turtlebot_frame} is the frame of the turtlebot. You can use the ar tag on top of the turtlebot as the frame or use a static transform publisher.
rosrun turtlebot_control turtlebot_control.py {color} {turtlebot_frame}
where {color} is the color of the turtlebot.
rosrun turtlebot_control turtlebot_control_lab_4.py