-
Notifications
You must be signed in to change notification settings - Fork 53
Description
- Velodyne VLP-16
I am using the official ROS driver:
https://github.com/ros-drivers/velodyne
Since the default point cloud format differs from the format expected by Simple-LIO-SAM, I made the following adjustments:
Set the point height to 1
Reordered the laser ring indices to match the ordering used in your provided sample dataset
I would like to confirm whether these modifications are correct and sufficient for compatibility with Simple-LIO-SAM.
- IMU (EGI-M320)
My IMU is the EGI-M320, which does not provide a magnetometer.
Its native frame has Z-axis pointing upward, and when the IMU is stationary:
acc_z ≈ +9.8 m/s²
acc_x ≈ 0
acc_y ≈ 0
I estimate roll and pitch from the accelerometer.
I would like to confirm the following regarding IMU data formatting:
(a) Should the IMU acceleration published to Simple-LIO-SAM remain positive 9.8 m/s² on the Z-axis when stationary, or should the sign be inverted?
(b) Should roll, pitch, and yaw angles be expressed relative to the ENU coordinate frame (X-east, Y-north, Z-up)?
(c) For roll/pitch/yaw:
When the IMU rotates counter-clockwise, should the angle value increase, or should it decrease, according to Simple-LIO-SAM’s convention?
Thank you very much for your time and assistance.
Any clarification regarding these coordinate frame conventions will be greatly appreciated.