| ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange |
I am currently developing a robot which will use a camera system to read an ArUco marker to give the robot its pose. The ArUco part is sorted, however I am having trouble building the navigationstack for this usecase. The robot will drive from point-to-ponit inputted by the user and current pose being fed back from the cameras.
Any tips or similar applications I can look for?
Thanks!
You can look into following packages: detection and navigation
Steps to follow:
Transform the detection. Since the detections you get are in different frame as what ROS uses so you need to transform them as follows
x_in_ros = z_detection
y_in_ros = - x_detection
z_in_ros = y_detection
Now you have distance of tag from the camera in ROS coordinates. You can give it as a goal using move_base action client or by just publishing to the goal topic of move_base.
You can also use the navigation link I have provided, then you don't need to change detection, you need to pass them as it is to navigation and it will publish velocity command to get you there.
Hope this helps :)
Asked: 2019-03-12 08:22:32 -0500
Seen: 530 times
Last updated: Mar 19 '19
ROS Answers is licensed under Creative Commons Attribution 3.0 Content on this site is licensed under a Creative Commons Attribution Share Alike 3.0 license.
