There seems to be an issue with sensor drift while my robot is rotating. I am not sure if it is because of the velodyne or the robot. I am using the Clearpath Jackal and mounting the Velodyne on top in Gazebo. When my robot moves forward and back, the laser points stay fairly in place (with some small movement). However, when the robot first starts accelerating in a rotation, the laser scan gains a considerable rotational drift in the same direction and retains that drift until the robot stops moving. Once it stops moving, it recovers back to it's original normal position.
I don’t know whether to blame the Velodyne simulator or the Jackal. It could have potentially to do with the Jackal odometry / and yaw estimates, but I am not sure. Does the Velodyne use the robot’s odometry in anyway to map points in simulation? Note I am working on Ubuntu 18.04 with ROS Melodic.
Any guidance or hints would be greatly appreciated.