REVISE: Robust Probabilistic Motion Planning in a Gaussian Random Field

Citation

If you use REVISE, please cite our paper (pdf)

@misc{rose2024reviserobustprobabilisticmotion,
      title={REVISE: Robust Probabilistic Motion Planning in a Gaussian Random Field}, 
      author={Alex Rose and Naman Aggarwal and Christopher Jewison and Jonathan P. How},
      year={2024},
      eprint={2411.13369},
      archivePrefix={arXiv},
      primaryClass={cs.RO},
      url={https://arxiv.org/abs/2411.13369}, 
}

General Setup

REVISE has been tested with Python 3.12.7 on MacOS and on Ubuntu 20.04.

Installing dependencies (using a virtual environment is recommended):

• pip install -r requirements.txt

Replicating paper results:

• python run_quadrotor_experiment.py to regenerate belief roadmaps for the single-query and multi-query experiments
• python run_monte_carlo_simulation.py to generate random goals for the multi-query experiment and simulate closed-loop trajectories for both experiments
• python run_metrics_evaluation.py to evaluate final state MSE, Wasserstein distance between the planned and actual final state distributions, and plan cost for both experiments
• python plot_results.py to regenerate the plots used in the paper

Documentation and Website

Documentation is auto-generated based on the source code and hosted by Read the Docs. Our project page is online at https://acl.mit.edu/REVISE/.

Acknowledgements

This research was supported by the National Science Foundation Graduate Research Fellowship under grant no. 2141064 and by the Draper Scholars program.

Website License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.