SQuAT Plan: Smooth Quadrotor Agile Trajectory Planning

SQuAT Plan is a Python-based framework for agile trajectory planning of quadrotors navigating through complex environments. It integrates nonlinear optimization (via GEKKO), obstacle avoidance, and both 3D and ROS-based visualizations.

Project Structure

squat-plan/
├── run.py                          # Unified entry point
├── pyproject.toml                  # Modern build system config
├── src/
│   └── squatplan/
│       ├── __init__.py
│       ├── main.py                 # Core simulation runner
│       ├── trajopt.py              # Trajectory optimization logic
│       ├── plotter.py              # Matplotlib-based plotting
│       ├── quaternion.py           # Quaternion math utils
│       ├── forester.py             # Obstacle generation
│       └── sphere_example_rviz.py  # ROS RViz marker publishing
├── squat.rviz                      # RViz display config
├── LICENSE
├── README.md
└── presentation.pdf                # MAE 271D presentation

Features

• Trajectory optimization using GEKKO with full or simplified dynamics
• Obstacle avoidance using geometric constraints
• Quaternion-based orientation modeling
• 3D visualizations via Matplotlib and RViz
• Synthetic forest generation for randomized path planning scenarios

Getting Started

Dependencies

Clone the repo and install in editable mode:

git clone https://github.com/aaronjohnsabu1999/squat-plan.git
cd squat-plan
python3 -m venv .venv
source venv/bin/activate
pip install -e .[dev]

Install ROS dependencies if on Linux or a WSL:

sudo apt install ros-${ROS_DISTRO}-rospy \
                 ros-${ROS_DISTRO}-geometry-msgs \
                 ros-${ROS_DISTRO}-visualization-msgs

Run Simulation

python run.py

To launch RViz in parallel:

roscore
# Then in another terminal:
python src/squatplan/sphere_example_rviz.py

Output Example

• Trajectory and state evolution plots (position, velocity, quaternion, thrust, moments)
• 3D environment with spherical/cylindrical obstacles and path trajectory

Project Context

Developed as a final project for MAE 271D — Control and Trajectory Planning for Autonomous Aerial Systems at UCLA.

Contributors:

• Aaron John Sabu
• Ryan Nemiroff
• Brett T. Lopez (Instructor)

Contact: {aaronjs, ryguyn, btlopez}@ucla.edu

License

MIT License © 2025
University of California, Los Angeles