OmniSpatialAI

All-in-One Spatial AI

OmniSpatialAI is a comprehensive repository dedicated to the field of Spatial AI, integrating Simultaneous Localization and Mapping (SLAM), Computer Vision, Control Systems, Geographic Information Systems (GIS), and Robotics (ROS).

---

🚀 Key Modules

• Control System: MATLAB/Simulink implementations for digital signal processing and control theory.
• SLAM: Advanced algorithms for Visual Odometry, Mapping, and Loop Closure, with benchmark datasets.
• GIS: Geographic data visualization and processing using Python and Folium.
• Planning: Path planning and mapping algorithms implemented in C++.
• ROS: Robot models (URDF/Xacro), sensor simulations (Gazebo), and actionlib tutorials.
• XR & Metaverse: Documentation and frameworks for AR/VR applications and tracking.

🛠 Tech Stack

• Languages: C++ (11/14/17), Python 3, MATLAB/Simulink.
• Frameworks: ROS (Noetic/Melodic), MkDocs.
• C++ Libraries: OpenCV, Eigen3, Sophus, Ceres, G2O, Pangolin.
• Python Libraries: Folium, NumPy, Matplotlib.

📁 Directory Structure

OmniSpatialAI/
├── control_system/     # Control theory & DSP (MATLAB)
│   └── basics/         # Transfer function & step response scripts
├── gis/                # Geographic Information Systems (Python)
├── planning/           # Path planning & Mapping (C++)
├── ros/                # ROS packages, URDFs, Gazebo simulations
│   └── extras/         # Non-catkin utilities (ros_matlab, ros_video, scripts)
├── slam/               # Visual Odometry, Mapping, SLAM frameworks
├── docs/               # Project documentation (MkDocs source)
└── mkdocs.yml          # Documentation configuration

📖 Documentation

The full documentation is available at sai.cgabc.xyz.

To run documentation locally:

pip install mkdocs-material
mkdocs serve

🏗 Getting Started

C++ Modules

mkdir build && cd build
cmake ..
make -j$(nproc)

ROS

Ensure you are in a catkin workspace:

catkin_make --source <path_to_omnispatialai>/ros/

📜 License

This project is licensed under the BSD 3-Clause License.