Situational Graphs - Reasoning

Situational Graphs - Reasoning is a ROS2 package for generating in real-time semantic concepts like Rooms and Walls from Wall Surfaces S-Graphs. For that purpose, Graph Neural Networks (GNNs) are used to estimate the existing relations between the wall surfaces.

📜 Table of contents

• 📖 Published Papers
• ⚙️ Installation
  • 📦 Installation with S-Graphs
  • 📦 Installation From Source
• 🚀 Usage
• ⚙️ Configuration files
• 🤖 ROS Related
  • 📥 Subscribed Topics
  • 📤 Published Topics

📖 Published Papers

Learning High-level Semantic-Relational Concepts for SLAM

⚙️ Installation

Note

Situational Graphs - Reasoning was only tested on Ubuntu 20.04, ROS2 Foxy, Humble Distros. We strongly recommend using cyclone_dds instead of the default fastdds.

📦 Installation with S-Graphs

Follow the S-Graphs installation instructions

📦 Installation From Source

Important

Before proceeding, make sure you have rosdep installed. You can install it using sudo apt-get install python3-rosdep In addition, ssh keys are needed to be configured on you GitHub account. If you haven't yet configured ssh keys, follow this tutorial

1. Update Rosdep:

rosdep init && rosdep update --include-eol-distros

2. Create a ROS2 workspace for S-Graphs

mkdir -p $HOME/workspaces && cd $HOME/workspaces

3. Clone the S-Graphs repository into the created workspace

git clone [email protected]:snt-arg/situational_graphs_reasoning.git -b develop

Important

If you have Nvidia GPU please install CUDA from this link. This code has only been tested with CUDA 11.8. If you dont have CUDA S-Graphs will use CPU only.

4. Install required dependencies. Change $ROS_DISTRO to your ros2 version.

cd situational_graphs_reasoning && source /opt/ros/$ROS_DISTRO/setup.sh && pip3 install -r requirements.txt

Note

If you want to compile with debug traces (from backward_cpp) run:

colcon build --symlink-install --cmake-args -DCMAKE_BUILD_TYPE=RelWithDebInfo

🚀 Usage

Follow the S-Graphs instructions to use this package along with all other functionalities.

Or launch situational_graphs_reasoning.py.

⚙️ Configuration files

File name	Description
config/same_room_training.json	Describes the data preprocessing and the GNN hyperparameters for room generation.
config/same_wall_training.json	Describes the data preprocessing and the GNN hyperparameters for wall generation.

🤖 ROS Related

📥 Subscribed Topics

situational_graphs_reasoning_node node

Topic name	Message Type	Description
/s_graphs/all_map_planes	s_graphs/PlanesData	Al the plains contained in the map.
/s_graphs/map_planes	s_graphs/PlanesData	Only the plains incorporated with the last keypoint.

📤 Published Topics

situational_graphs_reasoning_node node

Topic name	Message Type	Description
/room_segmentation/room_data	s_graphs/RoomsData	Contains all the necessary information about the rooms on a given floor.
/room_segmentation/wall_data	s_graphs/WallsData	Contains all the necessary information about the walls on a given floor.