We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
我在本项目中使用 mid40+ch229(imu)+ ros_noetic。但是发现:一开始标定还正常(整体还像个房间)。但三个都到100%后,无论是否做refine ,程序都认为静止的东西仍在运动(不仅仅是误差大,而且是整体上把静止当做运动的了)。(如下图,整个过程中我都在座位上)。我猜测可能是fastlio部分的原因?因为我在运行fastlio库的代码(完全没有外参标定)也出现了这种情况。lidar_imu_init 也是如此。
请问这大概是什么原因呢?
附上我的yaml
common: lid_topic: "/livox/lidar"
imu_topic: "/IMU_data"
preprocess: lidar_type: 1 # Livox series LiDAR feature_extract_en: false scan_line: 1 blind: 2
initialization: cut_frame_num: 5 # must be positive integer, 1 means do not cut frame orig_odom_freq: 10 # original Lidar frequency mean_acc_norm: 9.805 # 1 for livox built-in IMU online_refine_time: 0.0 # seconds data_accum_length: 400 Rot_LI_cov: [ 0.00005, 0.00005, 0.00005 ] Trans_LI_cov: [ 0.00001, 0.00001, 0.00001 ]
mapping: filter_size_surf: 0.05 filter_size_map: 0.15 gyr_cov: 50 acc_cov: 2 b_acc_cov: 0.0001 b_gyr_cov: 0.0001 det_range: 450.0
---- 后续更新 ---- 进行了室外环境的测量发现:标定的时候需要在3个方向上面有旋转和平移可以缓解上面的问题,这样至少不至于把静止的小车认为是在直线运动。但是任然存在下面问题:
The text was updated successfully, but these errors were encountered:
Were you able to solve this?
Sorry, something went wrong.
No branches or pull requests
我在本项目中使用 mid40+ch229(imu)+ ros_noetic。但是发现:一开始标定还正常(整体还像个房间)。但三个都到100%后,无论是否做refine ,程序都认为静止的东西仍在运动(不仅仅是误差大,而且是整体上把静止当做运动的了)。(如下图,整个过程中我都在座位上)。我猜测可能是fastlio部分的原因?因为我在运行fastlio库的代码(完全没有外参标定)也出现了这种情况。lidar_imu_init 也是如此。
请问这大概是什么原因呢?
附上我的yaml
common:
lid_topic: "/livox/lidar"
imu_topic: "/livox/imu/async"
preprocess:
lidar_type: 1 # Livox series LiDAR
feature_extract_en: false
scan_line: 1
blind: 2
initialization:
cut_frame_num: 5 # must be positive integer, 1 means do not cut frame
orig_odom_freq: 10 # original Lidar frequency
mean_acc_norm: 9.805 # 1 for livox built-in IMU
online_refine_time: 0.0 # seconds
data_accum_length: 400
Rot_LI_cov: [ 0.00005, 0.00005, 0.00005 ]
Trans_LI_cov: [ 0.00001, 0.00001, 0.00001 ]
mapping:
filter_size_surf: 0.05
filter_size_map: 0.15
gyr_cov: 50
acc_cov: 2
b_acc_cov: 0.0001
b_gyr_cov: 0.0001
det_range: 450.0
---- 后续更新 ----
进行了室外环境的测量发现:标定的时候需要在3个方向上面有旋转和平移可以缓解上面的问题,这样至少不至于把静止的小车认为是在直线运动。但是任然存在下面问题:
The text was updated successfully, but these errors were encountered: