This paper considers the integration of gap-based local navigation methods with artificial potential field (APF) methods to derive a local planning module for hierarchicalnavigation systems that has provable collision-free properties.Given that APF theory applies to idealized robot models, theprovable properties are lost when applied to more realistic models. We describe a set of algorithm modifications thatcorrect for these errors and enhance robustness to non-ideal models. Central to the construction of the local planner isthe use of sensory-derived local free-space models that detect gaps and use them for the synthesis of the APF. Modifications are given for a nonholonomic robot model. Integration of the local planner, called potential gap, into a hierarchical navigation system provides the local goals and trajectories needed for collision-free navigation through unknown environments.Monte Carlo experiments in benchmark worlds confirm the asserted safety and robustness properties by testing under various robot models.
[Demo Video], [Arxiv Preprint]
- Algorithm parameters
- Manuscript symbols
- Manuscript abbreviations
- Links to main implementation code files
- Algorithm details of gap simplification, radial gap conversion and radial extension
- Comparison Between TEB and PG
- Proof Files for Passage
- ROS (Kinetic Ubuntu 16.04) Installation Link
See NavBench https://github.com/ivalab/NavBench for rosinstall instructions and launching experiments.
@ARTICLE{9513583,
author={Xu, Ruoyang and Feng, Shiyu and Vela, Patricio},
journal={IEEE Robotics and Automation Letters},
title={Potential Gap: A Gap-Informed Reactive Policy for Safe Hierarchical Navigation},
year={2021},
volume={},
number={},
pages={1-1},
doi={10.1109/LRA.2021.3104623}
}
R. Xu, S. Feng and P. Vela, "Potential Gap: A Gap-Informed Reactive Policy for Safe Hierarchical Navigation," in IEEE Robotics and Automation Letters, doi: 10.1109/LRA.2021.3104623.
The source code is released under MIT license.