Note: This project is a work in progress.
A generalizable, automated scenario characterization framework for trajectory datasets.
Currently, this is a re-implementation of the scenario characterization approach introduced in SafeShift, as part of an internship project at StackAV.
Repository: github.com/navarrs/ScenarioCharacterization
This repository currently uses:
- uv as the package manager.
- Hydra for hierarchical configuration management.
- Pydantic for input/output data validation.
uv pip install scenario-characterization
Clone the repository and install the package in editable mode:
git clone [email protected]:navarrs/ScenarioCharacterization.git
cd ScenarioCharacterization
uv run pip install -e .To install with Waymo dependencies (required for running the example), use:
uv run pip install -e ".[waymo]"If installing with dev, run
uv run pip install -e. ".[dev]"
uv run pre-commit install- Organization: Overview of the Hydra configuration structure.
- Schemas: Guidelines for creating dataset adapters and processors that comply with the required input/output schemas.
- Characterization: Details on supported scenario characterization and visualization tools, and how to use them.
- Example: Step-by-step usage example using the Waymo Open Motion Dataset.
@INPROCEEDINGS{stoler2024safeshift,
author={Stoler, Benjamin and Navarro, Ingrid and Jana, Meghdeep and Hwang, Soonmin and Francis, Jonathan and Oh, Jean},
booktitle={2024 IEEE Intelligent Vehicles Symposium (IV)},
title={SafeShift: Safety-Informed Distribution Shifts for Robust Trajectory Prediction in Autonomous Driving},
year={2024},
volume={},
number={},
pages={1179-1186},
keywords={Meters;Collaboration;Predictive models;Robustness;Iron;Trajectory;Safety},
doi={10.1109/IV55156.2024.10588828}}
Run uv sync --frozen --all-groups to set up environment.
Run pre-commit run --all-files to run all hooks on all files.