Ask Your Question

karthik's profile - activity

2017-04-07 15:09:57 -0600 received badge  Popular Question (source)
2017-04-07 15:09:57 -0600 received badge  Notable Question (source)
2017-04-07 15:09:39 -0600 received badge  Notable Question (source)
2017-04-07 15:09:39 -0600 received badge  Popular Question (source)
2016-08-25 05:06:29 -0600 marked best answer /map and /odom after mapping the environment


1. I have managed to build a map using turtlebot simulator in gazebo and I see that the building of map is rather bad. I hope its because of the odometry instability in gazebo. Or am i missing something here. Any tricks to build a better map.

Note: I used the one that is available in turtlebot_navigation here.

2 I see that the tf's of /map and /odom after building the map are quite separated. Does this confirm my assumption that the odometry is not stable for the turtlebot simulation.

3 In that case can anyone suggest me a good robot model that has stable odometry.

Thanks, Karthik

Bad mapping: For example there are 4 tables in a room. When i move around the robot i see that after multiple scans of the same obstacle, the shape (occupied region) reduces to less than half and at times counts the actually occupied region to be free space.

2016-04-04 09:39:02 -0600 marked best answer What is an ideal way of debugging a segmentation fault in ROS?

Hi, I generally use gdb to debug my run time errors in my C++ programs. But in ros as it spins n waits for the msgs in topics, the gdb doesn't really work well. So I want to know if there is another method to debug the ros codes when they run into segmentation faults.

Thanks, Karthik

2016-04-04 09:38:49 -0600 received badge  Good Question (source)
2016-03-21 10:07:31 -0600 marked best answer Significance of ros::spinOnce()

Hi, Can somebody explain what is the actual significance of the ros::spinOnce(). How is it useful? What are the situations where its impact is higher? And also difference between ros::spin() ros::spinOnce().

Thanks, Karthik