IntelRealSense / librealsense

Intel® RealSense™ SDK
https://www.intelrealsense.com/
Apache License 2.0
7.63k stars 4.83k forks source link

The question about sdk and its corresponding FW version #13486

Closed dyb678 closed 6 days ago

dyb678 commented 2 weeks ago

Hi! I use the 2.19.0 sdk and I don't its corresponding FW version.Thanks!

MartyG-RealSense commented 2 weeks ago

Hi @dyb678 The recommended firmware version for SDK 2.19.0 is 5.11.4

https://github.com/IntelRealSense/librealsense/releases/tag/v2.19.0

image

However - and this is VERY important - only a D415 or D435 camera model could support a firmware that old. For example, if a D455 was updated with the 5.11.4 firmware then it would become unusable and PERMANENTLY no longer detectable.

image

dyb678 commented 2 weeks ago

OO,thanks!Actually,I find when I use the imu method" linear_interpolation" in ros wrapper will cause the data unstable but "copy"method seems not suitable for SLAM task. How can solve it ?Thanks a lot!

MartyG-RealSense commented 2 weeks ago

Raw RealSense IMU data is inherently "noisy". So there may not be much that you can do to stabilize the data if the Copy option cannot be used.

Ideally, fast moving or turning of the camera should be avoided if possible as this can cause the IMU's tracking to be lost.

You could also try setting accel and gyro to their maximum speed values to see how this affects the data when using linear_interpolation.

ROS1 roslaunch realsense2_camera rs_camera.launch enable_accel:=true enable_gyro:=true unite_imu_method:=linear_interpolation accel_fps:=200 gyro_fps:=400

ROS2 ros2 launch realsense2_camera rs_launch.py enable_accel:=true enable_gyro:=true unite_imu_method:=2 accel_fps:=200 gyro_fps:=400

MartyG-RealSense commented 1 week ago

Hi @dyb678 Do you require further assistance with this case, please? Thanks!

MartyG-RealSense commented 6 days ago

Case closed due to no further comments received.