SYNOPSIS

srba-slam --help

srba-slam --list-problems

srba-slam {--se2|--se3} {--lm-2d|--lm-3d} --obs [StereoCamera|...]
          -d DATASET.txt [--sensor-params-cfg-file SENSOR_CONFIG.cfg] 
          [--noise NOISE_SIGMA] [--verbose {0|1|2|3}] [--step-by-step]

COMPLETE LIST OF PARAMETERS

   srba-slam  [--eval-connectivity] [--eval-overall-se3-error]
              [--eval-overall-sqr-error] [--save-final-graph-landmarks
              <final-map.dot>] [--save-final-graph <final-map.dot>]
              [--debug-dump-cur-spantree] [--video-fps <>] [--gui-delay <>]
              [--create-video <out.avi>] [--cfg-file-rba-bootstrap
              <rba.cfg>] [--cfg-file-rba <rba.cfg>] [--random-seed <>] [-v
              <>] [--submap-size <20>] [--edge-policy
              <ecpICRA2013>] [--max-iters <>] [--max-lambda
              <depth>] [--max-optimize-depth <depth>]
              [--max-spanning-tree-depth <depth>] [--noise-ang <noise_std>]
              [--noise <noise_std>] [--add-noise] [--profile-stats-length
              <>] [--profile-stats <stats>] [--step-by-step] [--no-gui]
              [--list-problems] [--sensor-params-cfg-file <>] [--obs <>]
              [--graph-slam] [--lm-3d] [--lm-2d] [--se3] [--se2]
              [--max-fixed-feats-per-kf <>] [--gt-path <>] [--gt-map <>]
              [-d <>] [--] [--version] [-h]

DESCRIPTION

srba-slam is a command-line application for offline processing of robotic/vision datasets. See also the manual online: http://www.mrpt.org/srba

   --eval-connectivity
     At end, make stats on the graph connectivity
   --eval-overall-se3-error
     At end, evaluate the overall SE3 error for all relative poses
   --eval-overall-sqr-error
     At end, evaluate the overall square error for all the observations
     with the final estimated model
   --save-final-graph-landmarks <final-map.dot>
     Save the final graph-map (all KFs and all Landmarks) to a .dot file
   --save-final-graph <final-map.dot>
     Save the final graph-map of KFs to a .dot file
   --debug-dump-cur-spantree
     Dump to files the current spanning tree
   --video-fps <>
     If creating a video, its FPS (Hz).
   --gui-delay <>
     Milliseconds of delay between GUI frames. Default:0. Increase for
     correctly generating videos, etc.
   --create-video <out.avi>
     Creates a video with the animated GUI output (*.avi).
   --cfg-file-rba-bootstrap <rba.cfg>
     Writes an empty config file (*.cfg) for the RBA parameters and exit.
   --cfg-file-rba <rba.cfg>
     Config file (*.cfg) for the RBA parameters
   --random-seed <>
     <0: randomize; >=0, use this random seed.
   -v <>,  --verbose <>
     0:quiet, 1:informative, 2:tons of info
   --submap-size <20>
     Number of KFs in each 'submap' of the arc-creation policy.
   --edge-policy <ecpICRA2013>
     Policy for edge creation, as textual names of the enum
     TEdgeCreationPolicy
   --max-iters <>
     Max. number of optimization iterations.
   --max-lambda <depth>
     Marq-Lev. optimization: maximum lambda to stop iterating
   --max-optimize-depth <depth>
     Overrides this parameter in config files
   --max-spanning-tree-depth <depth>
     Overrides this parameter in config files
   --noise-ang <noise_std>
     One sigma of the noise model of every angular component of
     observations, in degrees (default: sensor-dependant)
     If a SRBA config is provided, it will override this value.
   --noise <noise_std>
     One sigma of the noise model of every component of observations
     (images,...) or to linear components if they're mixed (default:
     sensor-dependant)
     If a SRBA config is provided, it will override this value.
   --add-noise
     Add AWG noise to the dataset
   --profile-stats-length <>
     Length in KFs of each saved profiled segment
   --profile-stats <stats>
     Generate profile stats to CSV files, with the given prefix
   --step-by-step
     If showing the gui, go step by step
   --no-gui
     Don't show the live gui
   --list-problems
     List all implemented values for '--obs'
   --sensor-params-cfg-file <>
     Config file from where to load the sensor parameters
   --obs <>
     Type of observations in the dataset (use --list-problems to see
     available types)
   --graph-slam
     Define a relative graph-slam problem (no landmarks)
   --lm-3d
     Relative landmarks are Euclidean 3D points
   --lm-2d
     Relative landmarks are Euclidean 2D points
   --se3
     Relative poses are SE(3)
   --se2
     Relative poses are SE(2)
   --max-fixed-feats-per-kf <>
     Create fixed & known-location features
   --gt-path <>
     Ground-truth robot path file (e.g. 'dataset1_GT_PATH.txt', etc.)
   --gt-map <>
     Ground-truth landmark map file (e.g. 'dataset1_GT_MAP.txt', etc.)
   -d <>,  --dataset <>
     Dataset file (e.g. 'dataset1_SENSOR.txt', etc.)
   --,  --ignore_rest
     Ignores the rest of the labeled arguments following this flag.
   --version
     Displays version information and exits.
   -h,  --help
     Displays usage information and exits.

AUTHORS

srba-slam is part of the Mobile Robot Programming Toolkit (MRPT).

This manual page was written by Jose Luis Blanco <[email protected]>.

COPYRIGHT

This program is free software; you can redistribute it and/or modify it under the terms of the BSD License.

On Debian GNU/Linux systems, the complete text of the BSD License can be found in `/usr/share/common-licenses/BSD'.