feat(slam): add cartographer_ros

This commit is contained in:
X-lanni
2025-07-04 14:39:48 +08:00
parent df9cee5779
commit 5cf886315d
825 changed files with 92220 additions and 0 deletions
+24
View File
@@ -0,0 +1,24 @@
# Copyright 2016 The Cartographer Authors
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
set(OUTPUT_DIR "${PROJECT_BINARY_DIR}/docs/html")
add_custom_target(build_doc ALL
${SPHINX_EXECUTABLE} -b html
${CMAKE_CURRENT_SOURCE_DIR}/source
${CMAKE_CURRENT_BINARY_DIR}/html
COMMENT "Building documentation."
)
# TODO(hrapp): Install documentation?
Binary file not shown.

After

Width:  |  Height:  |  Size: 158 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 11 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 771 B

File diff suppressed because it is too large Load Diff

After

Width:  |  Height:  |  Size: 91 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 18 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.2 KiB

File diff suppressed because it is too large Load Diff

After

Width:  |  Height:  |  Size: 91 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.8 KiB

File diff suppressed because it is too large Load Diff

After

Width:  |  Height:  |  Size: 91 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 2.2 KiB

File diff suppressed because it is too large Load Diff

After

Width:  |  Height:  |  Size: 91 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 3.0 KiB

File diff suppressed because it is too large Load Diff

After

Width:  |  Height:  |  Size: 91 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 66 KiB

File diff suppressed because it is too large Load Diff

After

Width:  |  Height:  |  Size: 91 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.5 KiB

File diff suppressed because it is too large Load Diff

After

Width:  |  Height:  |  Size: 91 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 5.0 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 7.3 KiB

File diff suppressed because it is too large Load Diff

After

Width:  |  Height:  |  Size: 91 KiB

File diff suppressed because one or more lines are too long
@@ -0,0 +1,36 @@
<?xml version="1.0" encoding="utf-8"?>
<!-- Generator: Adobe Illustrator 19.0.0, SVG Export Plug-In . SVG Version: 6.00 Build 0) -->
<svg version="1.1" id="Layer_1" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px"
viewBox="-405 27 500 500" style="enable-background:new -405 27 500 500;" xml:space="preserve">
<style type="text/css">
.st0{fill:none;}
.st1{clip-path:url(#SVGID_2_);fill:#097138;}
.st2{clip-path:url(#SVGID_2_);fill:#0B8043;}
.st3{fill:#009688;}
.st4{fill:#33AC71;}
.st5{fill:#80CBC4;}
.st6{fill:#F5F5F5;}
</style>
<rect x="-395" y="37" class="st0" width="480" height="480"/>
<g>
<defs>
<path id="SVGID_1_" d="M65,277c0-60.8-24.6-115.8-64.4-155.6l0,0C-42.4,78.5-98.7,57-155,57v20c-55.2,0-105.2,22.4-141.4,58.6
l-14.1-14.1C-350.4,161.2-375,216.2-375,277s24.6,115.8,64.4,155.6l14.1-14.1c36.2,36.2,86.2,58.6,141.4,58.6v20
c60.8,0,115.8-24.6,155.6-64.4l-14.2-14.1C22.6,382.2,45,332.2,45,277H65z"/>
</defs>
<clipPath id="SVGID_2_">
<use xlink:href="#SVGID_1_" style="overflow:visible;"/>
</clipPath>
<circle class="st1" cx="-155" cy="277" r="200"/>
<path class="st2" d="M0.6,432.6C-39.2,472.4-94.2,497-155,497V277L0.6,432.6z"/>
<path class="st2" d="M-310.6,121.4C-350.4,161.2-375,216.2-375,277s24.6,115.8,64.4,155.6L-155,277L-310.6,121.4z"/>
<path class="st2" d="M0.6,121.4L0.6,121.4C-42.4,78.5-98.7,57-155,57v220H65C65,216.2,40.4,161.2,0.6,121.4z"/>
</g>
<path class="st3" d="M-34.8,156.8L-155,277l120.2,120.2C-4,366.5,15,324,15,277S-4,187.6-34.8,156.8z"/>
<path class="st4" d="M-34.8,156.8C-65.6,126-108.1,107-155,107s-89.4,19-120.2,49.8C-306,187.6-325,230.1-325,277
s19,89.5,49.8,120.2C-244.4,428-201.9,447-155,447s89.4-19,120.2-49.8L-155,277L-34.8,156.8z"/>
<path class="st5" d="M-155,207c-38.6,0-70,31.4-70,70s31.4,70,70,70c19.6,0,37.2-8,49.9-20.9l35.1,35.1c21.7-21.7,35-51.8,35-85
s-13.3-63.3-35-85l-35.9,35.9C-118.5,214.7-135.9,207-155,207z"/>
<path class="st6" d="M-155,347c-38.6,0-70-31.4-70-70s31.4-70,70-70c19.1,0,36.5,7.7,49.1,20.1l35.9-35.9c-21.7-21.7-51.8-35-85-35
c-66.2,0-120,53.8-120,120s53.8,120,120,120c33.2,0,63.3-13.3,85-35l-35.1-35.1C-117.8,339-135.4,347-155,347z"/>
</svg>

After

Width:  |  Height:  |  Size: 2.2 KiB

File diff suppressed because it is too large Load Diff

After

Width:  |  Height:  |  Size: 91 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 67 KiB

@@ -0,0 +1,6 @@
1: Light grey: Pantone Solid Uncoated 7541 U
2: Light teal: Pantone Solid Uncoated 564 U
3: Dark teal: Pantone Solid Coated 7473 C
4: Light green: Pantone Solid Coated 2250 C
5: Medium green: Pantone Solid Coated 7731 C
6: Dark green: Pantone Solid Coated 7732 C
Binary file not shown.

After

Width:  |  Height:  |  Size: 16 KiB

+275
View File
@@ -0,0 +1,275 @@
# -*- coding: utf-8 -*-
# Copyright 2016 The Cartographer Authors
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# Cartographer documentation build configuration file, created by
# sphinx-quickstart on Fri Jul 8 10:41:33 2016.
#
# This file is execfile()d with the current directory set to its
# containing dir.
#
# Note that not all possible configuration values are present in this
# autogenerated file.
#
# All configuration values have a default; values that are commented out
# serve to show the default.
import sys
import os
from datetime import datetime
# If extensions (or modules to document with autodoc) are in another directory,
# add these directories to sys.path here. If the directory is relative to the
# documentation root, use os.path.abspath to make it absolute, like shown here.
#sys.path.insert(0, os.path.abspath('.'))
# -- General configuration ------------------------------------------------
# If your documentation needs a minimal Sphinx version, state it here.
#needs_sphinx = '1.0'
# Add any Sphinx extension module names here, as strings. They can be
# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
# ones.
extensions = [
'sphinx.ext.todo',
'sphinx.ext.mathjax',
]
# Add any paths that contain templates here, relative to this directory.
templates_path = ['_templates']
# The suffix of source filenames.
source_suffix = '.rst'
# The encoding of source files.
#source_encoding = 'utf-8-sig'
# The master toctree document.
master_doc = 'index'
# General information about the project.
project = u'Cartographer'
copyright = u'{year} The Cartographer Authors'.format(year=datetime.now().year)
# The version info for the project you're documenting, acts as replacement for
# |version| and |release|, also used in various other places throughout the
# built documents.
#
# The short X.Y version.
#version = ''
# The full version, including alpha/beta/rc tags.
#release = ''
# The language for content autogenerated by Sphinx. Refer to documentation
# for a list of supported languages.
#language = None
# There are two options for replacing |today|: either, you set today to some
# non-false value, then it is used:
#today = ''
# Else, today_fmt is used as the format for a strftime call.
#today_fmt = '%B %d, %Y'
# List of patterns, relative to source directory, that match files and
# directories to ignore when looking for source files.
exclude_patterns = []
# The reST default role (used for this markup: `text`) to use for all
# documents.
#default_role = None
# If true, '()' will be appended to :func: etc. cross-reference text.
#add_function_parentheses = True
# If true, the current module name will be prepended to all description
# unit titles (such as .. function::).
#add_module_names = True
# If true, sectionauthor and moduleauthor directives will be shown in the
# output. They are ignored by default.
show_authors = False
# The name of the Pygments (syntax highlighting) style to use.
pygments_style = 'sphinx'
# A list of ignored prefixes for module index sorting.
#modindex_common_prefix = []
# If true, keep warnings as "system message" paragraphs in the built documents.
#keep_warnings = False
# -- Options for HTML output ----------------------------------------------
# The theme to use for HTML and HTML Help pages. See the documentation for
# a list of builtin themes.
html_theme = 'default'
# Theme options are theme-specific and customize the look and feel of a theme
# further. For a list of options available for each theme, see the
# documentation.
#html_theme_options = {}
# Add any paths that contain custom themes here, relative to this directory.
#html_theme_path = []
# The name for this set of Sphinx documents. If None, it defaults to
# "<project> v<release> documentation".
#html_title = None
# A shorter title for the navigation bar. Default is the same as html_title.
#html_short_title = None
# The name of an image file (relative to this directory) to place at the top
# of the sidebar.
#html_logo = None
# The name of an image file (within the static path) to use as favicon of the
# docs. This file should be a Windows icon file (.ico) being 16x16 or 32x32
# pixels large.
#html_favicon = None
# Add any paths that contain custom static files (such as style sheets) here,
# relative to this directory. They are copied after the builtin static files,
# so a file named "default.css" will overwrite the builtin "default.css".
html_static_path = []
# Add any extra paths that contain custom files (such as robots.txt or
# .htaccess) here, relative to this directory. These files are copied
# directly to the root of the documentation.
#html_extra_path = []
# If not '', a 'Last updated on:' timestamp is inserted at every page bottom,
# using the given strftime format.
#html_last_updated_fmt = '%b %d, %Y'
# If true, SmartyPants will be used to convert quotes and dashes to
# typographically correct entities.
#html_use_smartypants = True
# Custom sidebar templates, maps document names to template names.
#html_sidebars = {}
# Additional templates that should be rendered to pages, maps page names to
# template names.
#html_additional_pages = {}
# If false, no module index is generated.
#html_domain_indices = True
# If false, no index is generated.
#html_use_index = True
# If true, the index is split into individual pages for each letter.
#html_split_index = False
# If true, links to the reST sources are added to the pages.
#html_show_sourcelink = True
# If true, "Created using Sphinx" is shown in the HTML footer. Default is True.
#html_show_sphinx = True
# If true, "(C) Copyright ..." is shown in the HTML footer. Default is True.
#html_show_copyright = True
# If true, an OpenSearch description file will be output, and all pages will
# contain a <link> tag referring to it. The value of this option must be the
# base URL from which the finished HTML is served.
#html_use_opensearch = ''
# This is the file name suffix for HTML files (e.g. ".xhtml").
#html_file_suffix = None
# Output file base name for HTML help builder.
htmlhelp_basename = 'Cartographerdoc'
# -- Options for LaTeX output ---------------------------------------------
latex_elements = {
# The paper size ('letterpaper' or 'a4paper').
#'papersize': 'letterpaper',
# The font size ('10pt', '11pt' or '12pt').
#'pointsize': '10pt',
# Additional stuff for the LaTeX preamble.
#'preamble': '',
}
# Grouping the document tree into LaTeX files. List of tuples
# (source start file, target name, title,
# author, documentclass [howto, manual, or own class]).
latex_documents = [
('index', 'Cartographer.tex', u'Cartographer Documentation',
u'The Cartographer Authors', 'manual'),
]
# The name of an image file (relative to this directory) to place at the top of
# the title page.
#latex_logo = None
# For "manual" documents, if this is true, then toplevel headings are parts,
# not chapters.
#latex_use_parts = False
# If true, show page references after internal links.
#latex_show_pagerefs = False
# If true, show URL addresses after external links.
#latex_show_urls = False
# Documents to append as an appendix to all manuals.
#latex_appendices = []
# If false, no module index is generated.
#latex_domain_indices = True
# -- Options for manual page output ---------------------------------------
# One entry per manual page. List of tuples
# (source start file, name, description, authors, manual section).
man_pages = [
('index', 'cartographer', u'Cartographer Documentation',
[u'The Cartographer Authors'], 1)
]
# If true, show URL addresses after external links.
#man_show_urls = False
# -- Options for Texinfo output -------------------------------------------
# Grouping the document tree into Texinfo files. List of tuples
# (source start file, target name, title, author,
# dir menu entry, description, category)
texinfo_documents = [
('index', 'Cartographer', u'Cartographer Documentation',
u'The Cartographer Authors', 'Cartographer',
'Cartographer is a system that provides real-time simultaneous '
'localization and mapping (SLAM) in 2D and 3D across multiple platforms '
'and sensor configurations.', 'Miscellaneous'),
]
# Documents to append as an appendix to all manuals.
#texinfo_appendices = []
# If false, no module index is generated.
#texinfo_domain_indices = True
# How to display URL addresses: 'footnote', 'no', or 'inline'.
#texinfo_show_urls = 'footnote'
# If true, do not generate a @detailmenu in the "Top" node's menu.
#texinfo_no_detailmenu = False
+490
View File
@@ -0,0 +1,490 @@
.. Copyright 2016 The Cartographer Authors
.. Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
.. http://www.apache.org/licenses/LICENSE-2.0
.. Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
=============
Configuration
=============
.. DO NOT EDIT! This documentation is AUTOGENERATED, please edit .proto files as
.. needed and run scripts/update_configuration_doc.py.
cartographer.common.proto.CeresSolverOptions
============================================
bool use_nonmonotonic_steps
Configure the Ceres solver. See the Ceres documentation for more
information: https://code.google.com/p/ceres-solver/
int32 max_num_iterations
Not yet documented.
int32 num_threads
Not yet documented.
cartographer.mapping.pose_graph.proto.ConstraintBuilderOptions
==============================================================
double sampling_ratio
A constraint will be added if the proportion of added constraints to
potential constraints drops below this number.
double max_constraint_distance
Threshold for poses to be considered near a submap.
double min_score
Threshold for the scan match score below which a match is not considered.
Low scores indicate that the scan and map do not look similar.
double global_localization_min_score
Threshold below which global localizations are not trusted.
double loop_closure_translation_weight
Weight used in the optimization problem for the translational component of
loop closure constraints.
double loop_closure_rotation_weight
Weight used in the optimization problem for the rotational component of
loop closure constraints.
bool log_matches
If enabled, logs information of loop-closing constraints for debugging.
cartographer.mapping_2d.scan_matching.proto.FastCorrelativeScanMatcherOptions fast_correlative_scan_matcher_options
Options for the internally used scan matchers.
cartographer.mapping_2d.scan_matching.proto.CeresScanMatcherOptions ceres_scan_matcher_options
Not yet documented.
cartographer.mapping_3d.scan_matching.proto.FastCorrelativeScanMatcherOptions fast_correlative_scan_matcher_options_3d
Not yet documented.
cartographer.mapping_3d.scan_matching.proto.CeresScanMatcherOptions ceres_scan_matcher_options_3d
Not yet documented.
cartographer.mapping.pose_graph.proto.OptimizationProblemOptions
================================================================
double huber_scale
Scaling parameter for Huber loss function.
double acceleration_weight
Scaling parameter for the IMU acceleration term.
double rotation_weight
Scaling parameter for the IMU rotation term.
double local_slam_pose_translation_weight
Scaling parameter for translation between consecutive nodes based on the local SLAM pose.
double local_slam_pose_rotation_weight
Scaling parameter for rotation between consecutive nodes based on the local SLAM pose.
double odometry_translation_weight
Scaling parameter for translation between consecutive nodes based on the odometry.
double odometry_rotation_weight
Scaling parameter for rotation between consecutive nodes based on the odometry.
double fixed_frame_pose_translation_weight
Scaling parameter for the FixedFramePose translation.
double fixed_frame_pose_rotation_weight
Scaling parameter for the FixedFramePose rotation.
bool log_solver_summary
If true, the Ceres solver summary will be logged for every optimization.
cartographer.common.proto.CeresSolverOptions ceres_solver_options
Not yet documented.
cartographer.mapping.proto.MapBuilderOptions
============================================
bool use_trajectory_builder_2d
Not yet documented.
bool use_trajectory_builder_3d
Not yet documented.
int32 num_background_threads
Number of threads to use for background computations.
cartographer.mapping.proto.PoseGraphOptions pose_graph_options
Not yet documented.
cartographer.mapping.proto.MotionFilterOptions
==============================================
double max_time_seconds
Threshold above which range data is inserted based on time.
double max_distance_meters
Threshold above which range data is inserted based on linear motion.
double max_angle_radians
Threshold above which range data is inserted based on rotational motion.
cartographer.mapping.proto.PoseGraphOptions
===========================================
int32 optimize_every_n_nodes
Online loop closure: If positive, will run the loop closure while the map
is built.
cartographer.mapping.pose_graph.proto.ConstraintBuilderOptions constraint_builder_options
Options for the constraint builder.
double matcher_translation_weight
Weight used in the optimization problem for the translational component of
non-loop-closure scan matcher constraints.
double matcher_rotation_weight
Weight used in the optimization problem for the rotational component of
non-loop-closure scan matcher constraints.
cartographer.mapping.pose_graph.proto.OptimizationProblemOptions optimization_problem_options
Options for the optimization problem.
int32 max_num_final_iterations
Number of iterations to use in 'optimization_problem_options' for the final
optimization.
double global_sampling_ratio
Rate at which we sample a single trajectory's nodes for global
localization.
bool log_residual_histograms
Whether to output histograms for the pose residuals.
double global_constraint_search_after_n_seconds
If for the duration specified by this option no global contraint has been
added between two trajectories, loop closure searches will be performed
globally rather than in a smaller search window.
cartographer.mapping.proto.TrajectoryBuilderOptions
===================================================
cartographer.mapping_2d.proto.LocalTrajectoryBuilderOptions trajectory_builder_2d_options
Not yet documented.
cartographer.mapping_3d.proto.LocalTrajectoryBuilderOptions trajectory_builder_3d_options
Not yet documented.
bool pure_localization
Not yet documented.
cartographer.mapping_2d.proto.LocalTrajectoryBuilderOptions
===========================================================
float min_range
Rangefinder points outside these ranges will be dropped.
float max_range
Not yet documented.
float min_z
Not yet documented.
float max_z
Not yet documented.
float missing_data_ray_length
Points beyond 'max_range' will be inserted with this length as empty space.
int32 num_accumulated_range_data
Number of range data to accumulate into one unwarped, combined range data
to use for scan matching.
float voxel_filter_size
Voxel filter that gets applied to the range data immediately after
cropping.
cartographer.sensor.proto.AdaptiveVoxelFilterOptions adaptive_voxel_filter_options
Voxel filter used to compute a sparser point cloud for matching.
cartographer.sensor.proto.AdaptiveVoxelFilterOptions loop_closure_adaptive_voxel_filter_options
Voxel filter used to compute a sparser point cloud for finding loop
closures.
bool use_online_correlative_scan_matching
Whether to solve the online scan matching first using the correlative scan
matcher to generate a good starting point for Ceres.
cartographer.mapping_2d.scan_matching.proto.RealTimeCorrelativeScanMatcherOptions real_time_correlative_scan_matcher_options
Not yet documented.
cartographer.mapping_2d.scan_matching.proto.CeresScanMatcherOptions ceres_scan_matcher_options
Not yet documented.
cartographer.mapping.proto.MotionFilterOptions motion_filter_options
Not yet documented.
double imu_gravity_time_constant
Time constant in seconds for the orientation moving average based on
observed gravity via the IMU. It should be chosen so that the error
1. from acceleration measurements not due to gravity (which gets worse when
the constant is reduced) and
2. from integration of angular velocities (which gets worse when the
constant is increased) is balanced.
cartographer.mapping_2d.proto.SubmapsOptions submaps_options
Not yet documented.
bool use_imu_data
True if IMU data should be expected and used.
cartographer.mapping_2d.proto.RangeDataInserterOptions
======================================================
double hit_probability
Probability change for a hit (this will be converted to odds and therefore
must be greater than 0.5).
double miss_probability
Probability change for a miss (this will be converted to odds and therefore
must be less than 0.5).
bool insert_free_space
If 'false', free space will not change the probabilities in the occupancy
grid.
cartographer.mapping_2d.proto.SubmapsOptions
============================================
double resolution
Resolution of the map in meters.
int32 num_range_data
Number of range data before adding a new submap. Each submap will get twice
the number of range data inserted: First for initialization without being
matched against, then while being matched.
cartographer.mapping_2d.proto.RangeDataInserterOptions range_data_inserter_options
Not yet documented.
cartographer.mapping_2d.scan_matching.proto.CeresScanMatcherOptions
===================================================================
double occupied_space_weight
Scaling parameters for each cost functor.
double translation_weight
Not yet documented.
double rotation_weight
Not yet documented.
cartographer.common.proto.CeresSolverOptions ceres_solver_options
Configure the Ceres solver. See the Ceres documentation for more
information: https://code.google.com/p/ceres-solver/
cartographer.mapping_2d.scan_matching.proto.FastCorrelativeScanMatcherOptions
=============================================================================
double linear_search_window
Minimum linear search window in which the best possible scan alignment
will be found.
double angular_search_window
Minimum angular search window in which the best possible scan alignment
will be found.
int32 branch_and_bound_depth
Number of precomputed grids to use.
cartographer.mapping_2d.scan_matching.proto.RealTimeCorrelativeScanMatcherOptions
=================================================================================
double linear_search_window
Minimum linear search window in which the best possible scan alignment
will be found.
double angular_search_window
Minimum angular search window in which the best possible scan alignment
will be found.
double translation_delta_cost_weight
Weights applied to each part of the score.
double rotation_delta_cost_weight
Not yet documented.
cartographer.mapping_3d.proto.LocalTrajectoryBuilderOptions
===========================================================
float min_range
Rangefinder points outside these ranges will be dropped.
float max_range
Not yet documented.
int32 num_accumulated_range_data
Number of range data to accumulate into one unwarped, combined range data
to use for scan matching.
float voxel_filter_size
Voxel filter that gets applied to the range data immediately after
cropping.
cartographer.sensor.proto.AdaptiveVoxelFilterOptions high_resolution_adaptive_voxel_filter_options
Voxel filter used to compute a sparser point cloud for matching.
cartographer.sensor.proto.AdaptiveVoxelFilterOptions low_resolution_adaptive_voxel_filter_options
Not yet documented.
bool use_online_correlative_scan_matching
Whether to solve the online scan matching first using the correlative scan
matcher to generate a good starting point for Ceres.
cartographer.mapping_2d.scan_matching.proto.RealTimeCorrelativeScanMatcherOptions real_time_correlative_scan_matcher_options
Not yet documented.
cartographer.mapping_3d.scan_matching.proto.CeresScanMatcherOptions ceres_scan_matcher_options
Not yet documented.
cartographer.mapping.proto.MotionFilterOptions motion_filter_options
Not yet documented.
double imu_gravity_time_constant
Time constant in seconds for the orientation moving average based on
observed gravity via the IMU. It should be chosen so that the error
1. from acceleration measurements not due to gravity (which gets worse when
the constant is reduced) and
2. from integration of angular velocities (which gets worse when the
constant is increased) is balanced.
int32 rotational_histogram_size
Number of histogram buckets for the rotational scan matcher.
cartographer.mapping_3d.proto.SubmapsOptions submaps_options
Not yet documented.
cartographer.mapping_3d.proto.RangeDataInserterOptions
======================================================
double hit_probability
Probability change for a hit (this will be converted to odds and therefore
must be greater than 0.5).
double miss_probability
Probability change for a miss (this will be converted to odds and therefore
must be less than 0.5).
int32 num_free_space_voxels
Up to how many free space voxels are updated for scan matching.
0 disables free space.
cartographer.mapping_3d.proto.SubmapsOptions
============================================
double high_resolution
Resolution of the 'high_resolution' map in meters used for local SLAM and
loop closure.
double high_resolution_max_range
Maximum range to filter the point cloud to before insertion into the
'high_resolution' map.
double low_resolution
Resolution of the 'low_resolution' version of the map in meters used for
local SLAM only.
int32 num_range_data
Number of range data before adding a new submap. Each submap will get twice
the number of range data inserted: First for initialization without being
matched against, then while being matched.
cartographer.mapping_3d.proto.RangeDataInserterOptions range_data_inserter_options
Not yet documented.
cartographer.mapping_3d.scan_matching.proto.CeresScanMatcherOptions
===================================================================
double occupied_space_weight
Scaling parameters for each cost functor.
double translation_weight
Not yet documented.
double rotation_weight
Not yet documented.
bool only_optimize_yaw
Whether only to allow changes to yaw, keeping roll/pitch constant.
cartographer.common.proto.CeresSolverOptions ceres_solver_options
Configure the Ceres solver. See the Ceres documentation for more
information: https://code.google.com/p/ceres-solver/
cartographer.mapping_3d.scan_matching.proto.FastCorrelativeScanMatcherOptions
=============================================================================
int32 branch_and_bound_depth
Number of precomputed grids to use.
int32 full_resolution_depth
Number of full resolution grids to use, additional grids will reduce the
resolution by half each.
double min_rotational_score
Minimum score for the rotational scan matcher.
double min_low_resolution_score
Threshold for the score of the low resolution grid below which a match is
not considered. Only used for 3D.
double linear_xy_search_window
Linear search window in the plane orthogonal to gravity in which the best
possible scan alignment will be found.
double linear_z_search_window
Linear search window in the gravity direction in which the best possible
scan alignment will be found.
double angular_search_window
Minimum angular search window in which the best possible scan alignment
will be found.
cartographer.sensor.proto.AdaptiveVoxelFilterOptions
====================================================
float max_length
'max_length' of a voxel edge.
float min_num_points
If there are more points and not at least 'min_num_points' remain, the
voxel length is reduced trying to get this minimum number of points.
float max_range
Points further away from the origin are removed.
+130
View File
@@ -0,0 +1,130 @@
.. Copyright 2018 The Cartographer Authors
.. Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
.. http://www.apache.org/licenses/LICENSE-2.0
.. Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
==============
Cost functions
==============
Relative Transform Error 2D
===========================
Given two poses
:math:`\mathbf{p}_i = [\mathbf{x}_i; \theta_i] = [x_i, y_i, \theta_i]^T`
and :math:`\mathbf{p}_j = [\mathbf{x}_j; \theta_j] = [x_j, y_j, \theta_j]^T`
the transformation :math:`\mathbf T` from the coordinate frame :math:`j` to the
coordinate frame :math:`i` has the following form
.. math::
\mathbf{T}( \mathbf{p}_i,\mathbf{p}_j) =
\left[
\begin{array}{c}
R(\theta_i)^T (\mathbf x_j - \mathbf x_i) \\
\theta_j-\theta_i
\end{array}
\right]
where :math:`R(\theta_i)^T` is the rotation matrix of :math:`\theta_i`.
The weighted error :math:`f:\mathbb R^6 \mapsto \mathbb R^3` between
:math:`\mathbf T` and the measured transformation :math:`\mathbf T_{ij}^m =
[\mathbf x_{ij}^m; \theta_j^m]` from the coordinate frame :math:`j` to the
coordinate frame :math:`i` can be computed as
.. math::
\mathbf f_{\text{relative}}( \mathbf{p}_i,\mathbf{p}_j) =
\left[
w_{\text{t}} \; w_{\text{r}}
\right]
\left(
\mathbf T_{ij}^m - \mathbf T( \mathbf{p}_i,\mathbf{p}_j)
\right) =
\left[
\begin{array}{c}
w_{\text{t}}\left(
\mathbf x_{ij}^m - R(\theta_i)^T (\mathbf x_j - \mathbf x_i)
\right) \\
w_{\text{r}}\left(
\mathrm{clamp}(\theta_{ij}^m - (\theta_j-\theta_i))
\right)
\end{array}
\right]
where :math:`w_t` and :math:`w_r` are weights for translation and rotation
respectively and :math:`\mathrm{clamp}: \mathbb R \mapsto [-\pi, \pi]`
normalizes the angle difference.
Jacobian matrix :math:`J_f` is given by:
.. math::
\begin{align}
J_f( \mathbf{p}_i,\mathbf{p}_j) &=
\left[
\frac{\partial\mathbf f}{\partial x_i} \quad
\frac{\partial\mathbf f}{\partial y_i} \quad
\frac{\partial\mathbf f}{\partial \theta_i} \quad
\frac{\partial\mathbf f}{\partial x_j} \quad
\frac{\partial\mathbf f}{\partial y_j} \quad
\frac{\partial\mathbf f}{\partial \theta_j}
\right] \\
&\mathstrut \\
&=
\left[
\begin{array}{cccc}
w_{\text{t}} R^T(\theta_i)
& -w_{\text{t}} {\frac{\mathrm d R^T(\theta_i)}{\mathrm d \theta}}(\mathbf x_j - \mathbf x_i)
& -w_{\text{t}} R^T(\theta_i)
& \mathbf{0} \\
\mathbf{0}^T
& w_{\text{r}}
& \mathbf{0}^T
& -w_{\text{r}}
\end{array}
\right]
\end{align}
Landmark Cost Function
======================
Let :math:`\mathbf{p}_o` denote the global pose of the SLAM tracking frame at
which a landmark with the global pose :math:`\mathbf{p}_l` is observed.
The landmark observation itself is the measured transformation
:math:`\mathbf{T}^m_{ol}` that was observed at time :math:`t_o`.
As the landmark can be observed asynchronously, the pose of observation
:math:`\mathbf{p}_o` is modeled in between two regular, consecutive trajectory
nodes :math:`\mathbf{p}_i, \mathbf{p}_j`.
It is interpolated between :math:`\mathbf{p}_i` and
:math:`\mathbf{p}_j` at the observation time :math:`t_o` using a linear
interpolation for the translation and a quaternion SLERP for the rotation:
.. math::
\mathbf{p}_o = \text{interpolate}(\mathbf{p}_i, \mathbf{p}_j, t_o)
Then, the full weighted landmark cost function can be written as:
.. math::
\begin{align}
\mathbf f_{\text{landmark}}(\mathbf{p}_l, \mathbf{p}_i, \mathbf{p}_j) &=
\mathbf f_{\text{relative}}(\mathbf{p}_l, \mathbf{p}_o) \\
&=
\left[
w_{\text{t}} \; w_{\text{r}}
\right]
\left(
\mathbf T_{ol}^m - \mathbf T( \mathbf{p}_o,\mathbf{p}_l)
\right)
\end{align}
The translation and rotation weights :math:`w_{\text{t}}, w_{\text{r}}` are
part of the landmark observation data that is fed into Cartographer.
+109
View File
@@ -0,0 +1,109 @@
.. Copyright 2018 The Cartographer Authors
.. Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
.. http://www.apache.org/licenses/LICENSE-2.0
.. Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
==========
Evaluation
==========
Performing evaluation is a crucial part of developing a SLAM system.
For this purpose, Cartographer offers built-in tools that can aid the tuning process or can be used for quality assurance purposes.
These tools can be used to assess the SLAM result even when no dedicated ground truth is available.
This is in contrast to public SLAM benchmarks like e.g the KITTI dataset [1]_ or the TUM RGB-D dataset [2]_, where highly-precise ground truth states (GPS-RTK, motion capture) are available as a reference.
Concept
=======
The process comprises two steps:
1. auto-generation of "ground truth" relations
2. evaluation of the test data against the generated ground truth
The evaluation is based on the pose relations metric proposed in [3]_.
Rather than comparing the pose of a trajectory node directly to the corresponding ground truth pose, it compares the relative poses between two trajectory nodes in the probe data to the corresponding relation of two trajectory nodes in the ground truth trajectory.
In Cartographer, we can generate such ground truth relations from trajectories with loop closures.
Let an optimized trajectory with loop closures be the input for the ground truth generation.
We select the relations from loop closure constraints that satisfy the following criteria:
* ``min_covered_distance``: Minimum covered distance in meters before a loop closure is considered a candidate for autogenerated ground truth.
* ``outlier_threshold_meters``: Distance in meters beyond which constraints are considered outliers.
* ``outlier_threshold_radians``: Distance in radians beyond which constraints are considered outliers.
We can assume the pose relations of neighboring trajectory nodes fulfilling these requirements to be locally correct in a fully optimized trajectory.
Although this is not a ground truth in the sense of an independent input from another source, we can now use it to evaluate the quality of local SLAM results that were generated without loop closure optimization.
The following figure illustrates the concept.
On the left side, the ground truth relations are visualized as green connections between trajectory nodes of a fully optimized trajectory.
On the right side, the corresponding relations in a non-optimized trajectory are shown in red.
The actual metric that is computed is the difference between the ground truth (green) and the probe (red) relations.
.. editable version: https://drive.google.com/file/d/1riJCj8KK4tQZArssGRtoXt7aBr1SWEXM/view?usp=sharing
.. image:: autogenerate_groundtruth.png
Advantages & Limitations
========================
The first obvious advantage is the easier data collection process compared to a cumbersome ground truth setup.
Another great advantage of this methodology is that the SLAM system can be evaluated in any custom sensor configuration (compared to public benchmarks where we are restricted to the data and the sensor configuration of the authors).
However, this type of self-evaluation is not suitable for measuring the accuracy of the full SLAM system with all optimizations enabled - only an evaluation with *real* ground truth states can provide that.
Furthermore, trajectory nodes outside of loop closure areas can't be considered.
How-To
======
Given a serialized state of a fully optimized trajectory (here: ``optimized.pbstream`` file), the ground truth relations can be generated with the following command:
.. code-block:: bash
cd <build> # (directory where Cartographer's binaries are located)
./cartographer_autogenerate_ground_truth -pose_graph_filename optimized.pbstream -output_filename relations.pbstream -min_covered_distance 100 -outlier_threshold_meters 0.15 -outlier_threshold_radians 0.02
Then, a non-optimized trajectory ``test.pbstream`` can be evaluated against the generated relations with:
.. code-block:: bash
./cartographer_compute_relations_metrics -relations_filename relations.pbstream -pose_graph_filename test.pbstream
This will produce output in this form:
.. code-block:: none
Abs translational error 0.01944 +/- 0.01819 m
Sqr translational error 0.00071 +/- 0.00189 m^2
Abs rotational error 0.11197 +/- 0.12432 deg
Sqr rotational error 0.02799 +/- 0.07604 deg^2
----
References
==========
.. [1] Andreas Geiger, Philip Lenz and Raquel Urtasun.
*Are we ready for Autonomous Driving? The KITTI Vision Benchmark Suite*.
CVPR, 2012.
.. [2] Jürgen Sturm, Nikolas Engelhard, Felix Endres, Wolfram Burgard and Daniel Cremers.
*A Benchmark for the Evaluation of RGB-D SLAM Systems*.
IROS, 2012.
.. [3] Rainer Kümmerle, Bastian Steder, Christian Dornhege, Michael Ruhnke, Giorgio Grisetti, Cyrill Stachniss and Alexander Kleiner.
*On measuring the accuracy of SLAM algorithms.*
Autonomous Robots 27(4), pp.387-407, 2009.
Binary file not shown.

After

Width:  |  Height:  |  Size: 55 KiB

+121
View File
@@ -0,0 +1,121 @@
.. Copyright 2016 The Cartographer Authors
.. Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
.. http://www.apache.org/licenses/LICENSE-2.0
.. Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
============
Cartographer
============
.. toctree::
:maxdepth: 2
:hidden:
configuration
evaluation
terminology
cost_functions
pbstream_migration
`Cartographer`_ is a system that provides real-time simultaneous localization
and mapping (`SLAM`_) in 2D and 3D across multiple platforms and sensor
configurations.
.. _Cartographer: https://github.com/cartographer-project/cartographer
.. _SLAM: https://en.wikipedia.org/wiki/Simultaneous_localization_and_mapping
Technical Overview
==================
* High level system overview of Cartographer
.. image:: high_level_system_overview.png
:target: https://github.com/cartographer-project/cartographer/blob/master/docs/source/high_level_system_overview.png
.. To make modifications, edit the original Google Sketch and export a png.
.. https://docs.google.com/drawings/d/1kCJ_dEbSvV83THCUfMikCPw7xFrTkrvRw5r6Ji8C90c/edit?usp=sharing
Getting started
===============
Cartographer is a standalone C++ library. To get started quickly, use our `ROS
<http://www.ros.org>`_ integration.
Getting started with ROS
------------------------
ROS integration is provided by the `Cartographer ROS repository`_. You will find
complete documentation for using Cartographer with ROS at the
`Cartographer ROS Read the Docs site`_.
.. _Cartographer ROS repository: https://github.com/cartographer-project/cartographer_ros
.. _Cartographer ROS Read the Docs site: https://google-cartographer-ros.readthedocs.io
Getting started without ROS
---------------------------
Please see our ROS integration as a starting point for integrating your system
with the standalone library. Currently, it is the best available reference.
On Ubuntu 18.04 (Bionic):
.. literalinclude:: ../../scripts/install_debs_cmake.sh
:language: bash
:lines: 20-
.. literalinclude:: ../../scripts/install_abseil.sh
:language: bash
:lines: 20-
.. literalinclude:: ../../scripts/install_ceres.sh
:language: bash
:lines: 20-
.. literalinclude:: ../../scripts/install_proto3.sh
:language: bash
:lines: 20-
.. literalinclude:: ../../scripts/install_cartographer_cmake.sh
:language: bash
:lines: 20-
.. _system-requirements:
System Requirements
===================
Although Cartographer may run on other systems, it is confirmed to be working
on systems that meet the following requirements:
* 64-bit, modern CPU (e.g. 3rd generation i7)
* 16 GB RAM
* Ubuntu 18.04 (Bionic), 20.04 (Focal)
* gcc version 6.3.0, 7.5.0, 9.3.0
Known Issues
------------
* 32-bit builds have libeigen alignment problems which cause crashes and/or
memory corruptions.
How to cite us
==============
Background about the algorithms developed for Cartographer can be found in the
following publication. If you use Cartographer for your research, we would
appreciate it if you cite our paper.
W. Hess, D. Kohler, H. Rapp, and D. Andor,
`Real-Time Loop Closure in 2D LIDAR SLAM`_, in
*Robotics and Automation (ICRA), 2016 IEEE International Conference on*.
IEEE, 2016. pp. 12711278.
.. _Real-Time Loop Closure in 2D LIDAR SLAM: https://research.google.com/pubs/pub45466.html
@@ -0,0 +1,45 @@
.. Copyright 2018 The Cartographer Authors
.. Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
.. http://www.apache.org/licenses/LICENSE-2.0
.. Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
=================================
Migration tool for pbstream files
=================================
The pbstream serialization format for 3D has changed to include additional
data (histograms) in each submap. Code to load old data by migrating
on-the-fly will be removed soon. Once this happened, users who wish to
migrate old pbstream files can use a migration tool.
The tool is shipped as part of Cartographer's pbstream tool (`source`_) and once
built can be invoked as follows:::
cartographer_pbstream migrate old.pbstream new.pbstream
The tool assumes 3D data in the old submap format as input and converts it
to the currently used format version.
Migrating pre-1.0 pbstream files
================================
With the update of the pbstream serialization format as discussed in
`RFC-0021`_, previously serialized pbstream files are not loadable in
Cartographer 1.0 anymore.
In order to enable users to reuse previously generated pbstream files,
migration using an older version of the migration tool is necessary.
The current tool does not support this migration anymore. Please use
the version at Git SHA 6c889490e245cc5d9da15023249c6fc7119def3f.
.. _RFC-0021: https://github.com/cartographer-project/rfcs/blob/master/text/0021-serialization-format.md
.. _source: https://github.com/cartographer-project/cartographer/blob/master/cartographer/io/pbstream_main.cc
+71
View File
@@ -0,0 +1,71 @@
.. Copyright 2017 The Cartographer Authors
.. Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
.. http://www.apache.org/licenses/LICENSE-2.0
.. Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
===========
Terminology
===========
This documents a few common patterns that exist in the Cartographer codebase.
Frames
======
global map frame
This is the frame in which global SLAM results are expressed. It is the fixed
map frame including all loop closure and optimization results. The transform
between this frame and any other frame can jump when new optimization results
are available. Its z-axis points upwards, i.e. the gravitational acceleration
vector points in the -z direction, i.e. the gravitational component measured
by an accelerometer is in the +z direction.
local map frame
This is the frame in which local SLAM results are expressed. It is the fixed
map frame excluding loop closures and the pose graph optimization. For a given
point in time, the transform between this and the global map frame may change,
but the transform between this and all other frames does not change.
submap frame
Each submap has a separate fixed frame.
tracking frame
The frame in which sensor data is expressed. It is not fixed, i.e. it changes
over time. It is also different for different trajectories.
gravity-aligned frame
Only used in 2D. A frame colocated with the tracking frame but with a
different orientation that is approximately aligned with gravity, i.e. the
gravitational acceleration vector points approximately in the -z direction. No
assumption about yaw (rotation around the z axis between this and the tracking
frame) should be made. A different gravity-aligned frame is used for different
trajectory nodes, e.g. yaw can change arbitrarily between gravity-aligned
frames of consecutive nodes.
Transforms
==========
local_pose
Transforms data from the tracking frame (or a submap frame, depending on
context) to the local map frame.
global_pose
Transforms data from the tracking frame (or a submap frame, depending on
context) to the global map frame.
local_submap_pose
Transforms data from a submap frame to the local map frame.
global_submap_pose
Transforms data from a submap frame to the global map frame.