chore: remove unused plane segmentation and sensors

Author: Idate96

README.md

@@ -105,8 +105,9 @@ Inside docker container.
 ```zsh
 cd $HOME/catkin_ws
 catkin build elevation_mapping_cupy
-catkin build convex_plane_decomposition_ros  # If you want to use plane segmentation
-catkin build semantic_sensor  # If you want to use semantic sensors
+
+> [!NOTE]
+> The ``plane_segmentation`` and ``sensor_processing`` stacks were removed from this workspace because they are not maintained on ROS 2.  Use an older commit if you still need them.
 ```
 
 ### Run turtlebot example
@@ -118,28 +119,6 @@ export TURTLEBOT3_MODEL=waffle
 roslaunch elevation_mapping_cupy turtlesim_simple_example.launch
 ```
 
-For fusing semantics into the map such as rgb from a multi modal pointcloud:
-
-```bash
-export TURTLEBOT3_MODEL=waffle
-roslaunch elevation_mapping_cupy turtlesim_semantic_pointcloud_example.launch
-```
-
-For fusing semantics into the map such as rgb semantics or features from an image:
-
-```bash
-export TURTLEBOT3_MODEL=waffle
-roslaunch elevation_mapping_cupy turtlesim_semantic_image_example.launch
-```
-
-For plane segmentation:
-
-```bash
-catkin build convex_plane_decomposition_ros
-export TURTLEBOT3_MODEL=waffle
-roslaunch elevation_mapping_cupy turtlesim_plane_decomposition_example.launch
-```
-
 To control the robot with a keyboard, a new terminal window needs to be opened.
 Then run
 

TODO.md

@@ -1,11 +1,4 @@
 - [x] Compile "elevation_mapping_msgs" with ROS2
 - [x] Compile "elevation_mapping_cupy" with ROS2
-- [x] Compile "semantic_sensor" with ROS2
-- [x] Compile "convex_plane_decomposition_msgs" with ROS2
-- [ ] Compile "convex_plane_decomposition" with ROS2
-- [x] Compile "grid_map_filters_rsl" with ROS2
-- [x] Compile "cgal5_ament" with ROS2
 
 - [x] Run "elevation_mapping_cupy" with ROS2
-- [ ] Run "semantic_sensor" with ROS2
-- [ ] Run "convex_plane_decomposition" with ROS2

docker/build.sh

@@ -17,6 +17,3 @@ colcon build \
                 "-DBUILD_TESTING=OFF"\
                 "-DCMAKE_CXX_FLAGS="-Wl,--allow-shlib-undefined""\
         -Wall -Wextra -Wpedantic -Wshadow \
-        --packages-skip \
-                convex_plane_decomposition \
-                convex_plane_decomposition_ros

docs/source/conf.py

@@ -9,7 +9,6 @@
 import os, sys
 
 sys.path.append(os.path.join(os.path.dirname(__file__), '../../elevation_mapping_cupy/script'))
-sys.path.append(os.path.join(os.path.dirname(__file__), '../../sensor_processing/semantic_sensor/script'))
 
 autodoc_mock_imports = [
     "cupy",

docs/source/documentation.rst

@@ -20,8 +20,8 @@ Index
 
 
 This is a ROS package for elevation mapping on GPU. The elevation mapping code is written in python and uses cupy for GPU computation. The
-plane segmentation is done independently and runs on CPU. When the plane segmentation is generated, local convex approximations of the
-terrain can be efficiently generated.
+legacy plane segmentation pipeline ran independently on the CPU to provide convex terrain approximations, but it has been removed from the
+current ROS 2 workspace until it is validated again.
 
 .. image:: ../media/main_repo.png
     :alt: Elevation map examples

docs/source/getting_started/installation.rst

@@ -257,32 +257,6 @@ have the third-party libaries installed on you machine:
 
 Semantic Sensors
 ==================================================================
-Elevation mapping node can receive multi-modal point cloud and image topics.
-In this example, we use semantic segmentation models to process color images and publish those topics.
-
-Python dependencies
--------------------------------------------------------------------
-
-.. code-block:: bash
-
-  pip3 install scikit-learn
-
-Torchvision (for Jetson Orin on Ubuntu 20.04) 
-
-.. code-block:: bash
-
-  git clone --branch release/0.15 https://github.com/pytorch/vision torchvision
-  cd torchvision/
-  export BUILD_VERSION=0.15.1
-  python3 setup.py install --user
-
-Detectron
-
-.. code-block:: bash
-
-  python3 -m pip install 'git+https://github.com/facebookresearch/detectron2.git'
-
-
 Build
 ==================================================================
 After installing all the dependencies, you can build the packages.
@@ -293,5 +267,6 @@ Then, build the packages with catkin.
 
   cd <your_catkin_ws>
   catkin build elevation_mapping_cupy  # The core package
-  catkin build convex_plane_decomposition_ros  # If you want to use plane segmentation
-  catkin build semantic_sensor  # If you want to use semantic sensors
+
+.. note::
+   The legacy ``plane_segmentation`` and ``sensor_processing`` stacks have been removed from this workspace because they were untested on ROS 2.  Use an earlier revision if you rely on those packages.

docs/source/getting_started/introduction.rst

@@ -35,7 +35,7 @@ Key Features
 
 * **Learning-based Traversability Filter**: Assesses terrain traversability using local geometry, improving path planning and navigation.
 
-* **Versatile Locomotion Tools**: Incorporates smoothing filters and plane segmentation, optimizing movement across various terrains.
+* **Versatile Locomotion Tools**: Incorporates smoothing filters and, in legacy setups, plane segmentation to optimize movement across various terrains.
 
 * **Multi-Modal Elevation Map (MEM) Framework**: Allows seamless integration of diverse data like geometry, semantics, and RGB information, enhancing multi-modal robotic perception.
 
@@ -78,8 +78,6 @@ Example setup is in **elevation_mapping_cupy/config/core/example_setup.yaml**.
 
   The transformation tree.
 
-* The plane segmentation node subscribes to an elevation map topic ([grid_map_msg/GridMap]). This can be configured in
-  **convex_plane_decomposition_ros/config/core/parameters.yaml**
 
 Published Topics
 -------------------------------------------------------------------
@@ -110,23 +108,3 @@ Example setting in `config/parameters.yaml`.
 * **elevation_map_filter** ([grid_map_msg/GridMap])
 
   The filtered maps using plugins.
-
-The plane segmentation node publishes the following:
-
-* **planar_terrain**  ([convex_plane_decomposition_msgs/PlanarTerrain])
-
-  A custom message that contains the full segmentation as a map together with the boundary information.
-
-* **filtered_map**  ([grid_map_msg/GridMap])
-
-  A grid map message to visualize the segmentation and some intermediate results. This information is also part of **planar_terrain**.
-
-* **boundaries**  ([visualization_msgs/MarkerArray])
-
-  A set of polygons that trace the boundaries of the segmented region. Holes and boundaries of a single region are published as separate
-  markers with the same color.
-
-* **insets**  ([visualization_msgs/PolygonArray])
-
-  A set of polygons that are at a slight inward offset from **boundaries**. There might be more insets than boundaries since the inward
-  shift can cause a single region to break down into multiple when narrow passages exist.

docs/source/getting_started/tutorial.rst

@@ -2,8 +2,8 @@
 
 Tutorial
 ******************************************************************
-This tutorial will guide you through the basic usage of the elevation mapping cupy. 
-You will learn how to run the plane segmentation node, the sensor node, and the TurtleBot example. 
+This tutorial will guide you through the basic usage of elevation_mapping_cupy.
+You will learn how to run the core node and the TurtleBot example.
 
 If you want to use your own custom plugins, please refer to the :ref:`plugins`.
 
@@ -18,20 +18,6 @@ Basic usage.
   roslaunch elevation_mapping_cupy elevation_mapping_cupy.launch
 
 
-For the plane segmentation node
-
-.. code-block:: bash
-
-  roslaunch convex_plane_decomposition_ros convex_plane_decomposition.launch
-
-For the sensor node
-
-.. code-block:: bash
-
-  roslaunch semantic_sensor semantic_pointcloud.launch
-  roslaunch semantic_sensor semantic_image.launch
-
-
 Run TurtleBot example
 ==================================================================
 
@@ -55,27 +41,6 @@ For fusing semantics into the map such as rgb from a multi modal pointcloud:
 .. image:: ../../media/turtlebot.png
     :alt: Elevation map examples
 
-.. code-block:: bash
-
-  export TURTLEBOT3_MODEL=waffle
-  roslaunch elevation_mapping_cupy turtlesim_semantic_pointcloud_example.launch
-
-For fusing semantics into the map such as rgb semantics or features from an image:
-
-.. code-block:: bash
-
-  export TURTLEBOT3_MODEL=waffle
-  roslaunch elevation_mapping_cupy turtlesim_semantic_image_example.launch
-
-Or, for the version including plane segmentation:
-
-.. code-block:: bash
-
-  catkin build convex_plane_decomposition_ros
-  export TURTLEBOT3_MODEL=waffle
-  roslaunch elevation_mapping_cupy turtlesim_plane_decomposition_example.launch
-
-
 To control the robot with a keyboard, a new terminal window needs to be opened.
 Then run
 
@@ -87,6 +52,8 @@ Then run
 
 Velocity inputs can be sent to the robot by pressing the keys `a`, `w`, `d`, `x`. To stop the robot completely, press `s`.
 
+.. note::
+   The previous semantic sensor and plane segmentation demos were removed from this workspace because they have not been validated on ROS 2.  Use a historical revision if you still need them.
 
 
 
@@ -124,4 +91,3 @@ Build with option.
   catkin build elevation_mapping_cupy -DPYTHON_EXECUTABLE=$(which python3)
 
 
-

docs/source/index.rst

@@ -21,7 +21,6 @@
 
      Plugins <usage/plugins>
      Parameters <usage/parameters>
-     Plane Segmentation <usage/plane_segmentation>
      Semantics <usage/semantics>
 
 
@@ -36,4 +35,3 @@
 
 
 
-

docs/source/python/index.rst

@@ -3,6 +3,4 @@ Python software
 
 .. toctree::
   elevation_mapping
-  semantic_sensor
-