IMU messages (sensor_msgs/Imu Message) using incorrect units?

[greymfm]

According to ROS docs, the IMU messages should use m/s² units (and not g's) for acceleration etc.:

http://docs.ros.org/en/noetic/api/sensor_msgs/html/msg/Imu.html

In the 'fake_imu' code, I don't see any earth gravity constant (1g = 9.8 m/s²) in the acceleration computation:

mcl_3dl/src/mcl_3dl.cpp

Line 241 in 1ebf897

imu->linear_acceleration.x = accel.x_;

Maybe 'mcl_3dl' is using wrong units in the expected IMU messages (e.g. IMU callbacks)? :-)

mcl_3dl/src/mcl_3dl.cpp

Line 942 in 1ebf897

void cbImu(const sensor_msgs::Imu::ConstPtr& msg)

Simulating ROS IMU messages should probably look something like this (according to ROS specifications)? :-)
imu->linear_acceleration.x = accel.x_ * 9.8;
imu->linear_acceleration.y = accel.y_ * 9.8;
imu->linear_acceleration.z = accel.z_ * 9.8;

[at-wat]

mcl_3dl node only uses vector direction of the linear acceleration, so absolute length of the vector doesn't matter.

[greymfm]

Thanks for clarification!