Metadata-Version: 2.4
Name: sentry-uav
Version: 0.0.0
Summary: Name reservation placeholder for sentry-uav (SENTRY UAV trajectory prediction toolkit). Real release is forthcoming with the SoftwareX paper.
Author-email: Sandeco <sanderson.macedo@ifg.edu.br>
Maintainer-email: Sandeco <sanderson.macedo@ifg.edu.br>
License-Expression: MIT
Project-URL: Documentation, https://github.com/sandeco/sentry-uav
Project-URL: Source, https://github.com/sandeco/sentry-uav
Project-URL: Issues, https://github.com/sandeco/sentry-uav/issues
Keywords: uav,drone,trajectory-prediction,computer-vision,calibrated-uncertainty,sentry
Classifier: Development Status :: 1 - Planning
Classifier: Intended Audience :: Science/Research
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Topic :: Scientific/Engineering :: Image Recognition
Classifier: Typing :: Typed
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Dynamic: license-file

# sentry-uav

Placeholder for the upcoming `sentry-uav` Python package.

**Status:** name reservation only. Version `0.0.0` exists exclusively to secure the project name on PyPI ahead of the first functional release.

## What is this?

`sentry-uav` will be the open source distribution of the **SENTRY** trajectory prediction system for unmanned aerial vehicles, built on two specialist Transformers:

- **SIGNAL**, the deterministic signature driven linear forecaster.
- **PROBE**, the probabilistic bivariate ellipse forecaster with calibrated 95 % confidence regions.

The package will provide a clean Python API and a Typer based CLI, distribute pretrained checkpoints via `from_pretrained()`, and reproduce the results published in the SENTRY paper (artigo 2 da tese).

## When is the real release?

Version `0.1.0` will land together with the SoftwareX submission of the companion paper. Until then, do not install version `0.0.0` for any real workload.

## References

- SENTRY paper (in preparation), Elsevier *Array*.
- Drone detection in airport environments, a literature review, *Array* 28 (2025) 100511, DOI 10.1016/j.array.2025.100511.

## License

MIT, see `LICENSE`.
