This project tackles the challenge of turning raw ship location data into full vessel trajectories, especially when vital identification numbers like IMO or MMSI are missing. Our track_builder Python module infers these continuous ship tracks. It does this by analyzing how close individual data points are to each other in time and space, then uses other available details—like a ship's speed, course, or type—to make connections. In essence, this algorithm reconstructs ship trajectories from scattered, unidentified data points by linking them based on where and when they occurred, along with their associated attributes.
- Identifier-Agnostic Trajectory Reconstruction: Connects ship points into tracks even when unique vessel identifiers (like MMSI/IMO) aren't present.
- Spatiotemporal Linking: Uses geographic location and timestamps to figure out likely connections between data points.
- Attribute-Based Validation: Incorporates additional data attributes, such as changes in speed or heading, to confirm or refine point associations.
- Scalable Approach: Designed to efficiently process large volumes of Arctic shipping traffic data.
You can install the track_builder module directly from this repository.
git clone git@github.com:uga-gaim/TrackBuilder.git
cd TrackBuilderIf you only want to use the module in your projects, install it using pip:
pip install .If you plan to contribute to the project, run tests, or use development tools like pytest, install the module along with its development dependencies:
# [Option] Manage a dedicated environment for dev
# conda create -n venv_ship python==3.12 -y
# conda activate venv_ship
# conda install -c conda-forge jupyterlab ipywidgets tqdm ipykernel -y
pip install -e ".[dev]"We welcome contributions! If you have suggestions for improvements, spot any bugs, or want to contribute code, please feel free to open an issue or submit a pull request.
This project is licensed under the MIT License—you can find more details in the LICENSE file.