Today we added a CHR UM6-LT IMU to the omni bot communications is through a sparkfun FTDI board
the first problem we encountered was the RX and TX line need to be reversed. The other problem was
quanterions need to be enabled which can only be done in windows.
After downloading the ROS package imu_um6 (searching ROS under sensors not IMU) and
launching we were publishing data from the IMU. Next step is to add a covariance matrix and run robot pose EKF
Hey peter i was looking at using the um6 for my imu. does it actually work well? there isn't much documentation on how to use the imu. what does the co matrix do for you and how does it work. i thought in the example code it published out everything that was needed.
ReplyDeleteJohn
ReplyDeleteI haven't had much luck in running the IMU with EKF_filter It works you can see it as the pink arrow in the video above but when i use it with robot_efk filter i get transform errors. The co matrix is still a bit of mystery but it wont work with out one the numbers depend on the encoder errors per cycle there is a write up in
http://answers.ros.org/question/12642/calculating-covariances-of-robots-odometry
I'm very early in the development cycle of both odometery and the IMU and learning as i go along but when i find the answers i will publish them here
John
ReplyDeleteI have the IMU and EKF working and providing a good localisation doing a loop around the lounge and kitchen i am out by about 1 meter in Y and .5 meter in X
the pose angle is 90 degrees out (tuning localisation helps this) when you compare this to raw odometery of 12 meters in x and .5 in Y