Latest¶

Select keyframes from a SLAM trajectory and interpolate the corresponding reference positions, either from per-position txt files or from a ROS 2 bag (NavSatFix or Odometry messages).
Usage: keyframe_interpolation [OPTIONS] positions-path poses-path [out-dir]

Positionals:
  positions-path TEXT REQUIRED
                              Reference-data source. Auto-detected: a single .txt file is read line-by-line ("stamp x y z x_stddev y_stddev z_stddev"); a directory of per-position .txt files (filename = "<sec>_<nanosec>.txt") is read as such; a .mcap / .db3 / .sqlite3 file or a ROS 2 bag directory is read as a bag.
  poses-path TEXT:FILE REQUIRED
                              Path to SLAM trajectory in GLIM format
  out-dir TEXT [.]            Output directory for poses_keyframes.txt, positions_interpolated.txt and keyframe_interpolation.yaml (effective config including the ENU origin)

Options:
  -h,--help                   Print this help message and exit


Bag input:
  --pos-topic TEXT            Topic name (NavSatFix or Odometry) — required when positions-path is a bag
  -t,--target-frame TEXT      Frame of the SLAM poses (e.g. base_link). Reference positions are moved into this frame with the vehicle orientation and lever arm from /tf and /tf_static of the bag (full TF chain required; for NavSatFix the TF root must be ENU-aligned and connected through a moving /tf transform)


Origin:
  --origin FLOAT FLOAT FLOAT x 3
                              Origin of the local ENU frame [lat lon alt], alt above mean sea level. NavSatFix defaults to the first accepted fix; txt input is read as lat lon ellipsoidal-alt if given, as Cartesian coordinates otherwise


Algorithm:
  --stddev-threshold FLOAT [0.25] 
                              Reject reference frames whose horizontal stddev exceeds this value
  --keyframe-delta-x FLOAT [2] 
                              Minimum translational distance between keyframes [m]
  --keyframe-delta-angle FLOAT [0.5] 
                              Minimum rotational distance between keyframes [rad]
  --interpolate,--no-interpolate{false}
                              Spline-interpolate the reference position at each keyframe timestamp (otherwise use closest-neighbor selection)
  --interp-pos-delta-xyz FLOAT [0.25] 
                              Minimum euclidean distance between consecutive spline support points [m]
  --max-time-diff FLOAT:POSITIVE [0.5] 
                              Maximum time [s] between a keyframe and the nearest reference position (on both sides when interpolating); other keyframes are skipped. The reference data needs a rate of at least 1 / max-time-diff


Examples (positions-path auto-detected):
  # single txt file with one position per line
  flexcloud-keyframe-interpolation positions.txt poses_GLIM.txt

  # directory of per-position txt files
  flexcloud-keyframe-interpolation /path/to/positions/ poses_GLIM.txt

  # ROS 2 bag with NavSatFix on /sensor/gnss/fix
  flexcloud-keyframe-interpolation /path/to/bag.mcap poses_GLIM.txt /path/to/out \
      --pos-topic /sensor/gnss/fix --origin 48.262 11.667 0.0 --target-frame base_link

  # ROS 2 bag with Odometry, moved into the SLAM pose frame via TF
  flexcloud-keyframe-interpolation /path/to/bag.mcap poses_GLIM.txt /path/to/out \
      --pos-topic /odom --target-frame base_link

Georeference a SLAM trajectory and (optionally) a corresponding point cloud map by aligning it to a GNSS / reference trajectory using Umeyama and rubber-sheeting.
Usage: georeferencing [OPTIONS] positions-path poses-path

Positionals:
  positions-path TEXT:FILE REQUIRED
                              Path to GNSS / reference trajectory
  poses-path TEXT:FILE REQUIRED
                              Path to SLAM trajectory in GLIM format

Options:
  -h,--help                   Print this help message and exit


Inputs:
  --pcd TEXT                  Optional point cloud map to transform alongside the trajectory
  --config-file TEXT:FILE     Optional YAML file with index-based fine-tuning arrays (exclude_ind, shift_ind, shift_ind_dist, fake_ind, fake_ind_dist, fake_ind_height)


Trajectory matching:
  --control-points INT:POSITIVE [10] 
                              Number of control points for rubber-sheeting
  --stddev-threshold FLOAT [0.25] 
                              Maximum stddev of reference points for automatic control-point selection
  --square-size FLOAT FLOAT FLOAT [[0.1,0.1,10]]  x 3
                              Padding of enclosing square around trajectories [x y z] (fractions)


Origin:
  --origin FLOAT FLOAT FLOAT [{}]  x 3
                              Custom ENU zero point [lat lon alt], alt above mean sea level; the positions are then read as lat lon ellipsoidal-alt


Evaluation:
  --evaluation                Print trajectory-matching statistics (RMSE, mean, median, stddev of GNSS-vs-aligned and GNSS-vs-rubber-sheeted deviations) to the terminal and log per-segment, deviation-colored linestrings to the rerun viewer.


Examples:
  # cartesian reference, no point cloud, default parameters
  ros2 run flexcloud georeferencing positions_interpolated.txt poses_keyframes.txt

  # GPS reference, custom origin, transform a point cloud as well
  ros2 run flexcloud georeferencing reference.txt poses_keyframes.txt \
      --pcd map.pcd \
      --origin 48.262 11.667 0.0

  # supply index-based fine-tuning arrays via YAML
  ros2 run flexcloud georeferencing reference.txt poses_keyframes.txt \
      --config-file georeferencing.yaml