How to capture IMU dara from the LiDAR?

roya

Hello

I want to capture the real-time IMU data from the lidar. I know there is a "get_imu_intrinsics" code to capture the IMU data but I am not sure it is the real-time data or just the IMU configuration.
Could you please help me with this issue?

Many thanks in advance
Roya

BToschi

Hi roya,

IMU data is streamed to the configured port of your machine via UDP, by default it's port 7502 for Lidar packets and port 7503 for IMU packets. I'm not sure about ROS solution, but can tell you that Ouster Studio does not yet capture IMU packets, which is (besides performance) the major reason I wrote (so far: hacked) my own capture program with Qt/C++. Sorry, but I cannot share the code in current state, but it was straight-forward following the SW user guide.

The "imu_intrinsics" are important if you want to translate IMU readings to global Lidar coordinate system - the IMU is not placed at center of Lidar. For my Lidar it's a translation by (6.253 ; -11.775 ; 7.645) .

I hope this helps you getting further,
Best,
Bjoern

Choong

BToschi Hello, BToschi,

I have some trouble with receiving IMU data while using OS2-128.
In Ouster Studio, I can see arbitrary large integers through sensor stream in Outer Studio but I cannot figure out what the integers exactly mean.
According to the User Guide, it decribes only that the sensor sends out acceleration in g and angular velocity in deg/sec by '32bit float format'. I think some processing is needed to get real data.
Is there any scale factor that the large integer output should be devided by it?
Otherwise, How can I get real acceleration and angular velocity with the large value.

Here is my c-code as below.

pu8Buff[]: unsiged char IMU Data Packet.
iPosOfBuff : index of pu8Buff[]

AccelX = (float32_t)(pu8Buff[iPosOfBuff + 27] << 24 | pu8Buff[iPosOfBuff + 26] << 16 | pu8Buff[iPosOfBuff + 25] << 8 | pu8Buff[iPosOfBuff + 24] / 16384);//  g
AccelY = (float32_t)(pu8Buff[iPosOfBuff + 31] << 24 | pu8Buff[iPosOfBuff + 30] << 16 | pu8Buff[iPosOfBuff + 29] << 8 | pu8Buff[iPosOfBuff + 28]);//  g
AccelZ = (float32_t)(pu8Buff[iPosOfBuff + 35] << 24 | pu8Buff[iPosOfBuff + 34] << 16 | pu8Buff[iPosOfBuff + 33] << 8 | pu8Buff[iPosOfBuff + 32]);//  g
AngularVelX = (float32_t)(pu8Buff[iPosOfBuff + 39] << 24 | pu8Buff[iPosOfBuff + 38] << 16 | pu8Buff[iPosOfBuff + 37] << 8 | pu8Buff[iPosOfBuff + 36]);//  deg/sec
AngularVelY = (float32_t)(pu8Buff[iPosOfBuff + 43] << 24 | pu8Buff[iPosOfBuff + 42] << 16 | pu8Buff[iPosOfBuff + 41] << 8 | pu8Buff[iPosOfBuff + 40]);//  deg/sec
AngularVelZ = (float32_t)(pu8Buff[iPosOfBuff + 47] << 24 | pu8Buff[iPosOfBuff + 46] << 16 | pu8Buff[iPosOfBuff + 45] << 8 | pu8Buff[iPosOfBuff + 44]);//  deg/sec

I hope I could get some advice.
Thanks.

BToschi

Hi Choong,

sorry, so far I never tried to decode the IMU packets, I have other priorities.
I'd suggest to check endianess and position in stream first, when I wrote my code I also cross-checked with ROS driver, which may also give you clues how to continue for IMU packets.

Best regards,
Bjoern