| ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange |
Hi there. I m making mapping with my turtlebot by using ros gmapping. I notice that the slower the speed of turtlebot movement, the better map result that i got in rvis map. But i dont know why, and what theory that can support me to say this.please help me by giving the best reason.
regards
gmapping is CPU-intensive, particularly in configurations like the turtlebot, where it needs to run updates frequently to compensate for the poor odometry and poor field of view of the Kinect (compared to a scanning laser). The updates for gmapping are performed either when the robot has turned by more than a specified amount, or when the robot has moved linearly by more than a specified amount.
I suspect what's happening is that update step in gmapping is taking a long time on your CPU, and it isn't able to update as frequently as the parameters specify. By driving the robot more slowly, you give it more time to compute between updates.
To check this, you can watch the CPU usage on the turtlebot while driving it around, and you can watch the log output from gmapping, to see how frequently it's performing updates and resampling.
Asked: 2014-12-14 16:07:11 -0500
Seen: 285 times
Last updated: Dec 14 '14
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.
Are you running gmapping on the turtlebot while you drive it around, or are you running gmapping afterwards, on a bag file?
Im running the gmapping while drive it around