@misc{uwacu2023hierarchical,
title={Hierarchical Annotated Skeleton-Guided Tree-based Motion Planning},
author={Diane Uwacu and Ananya Yammanuru and Keerthana Nallamotu and Vasu Chalasani and Marco Morales and Nancy M. Amato},
year={2023},
eprint={2309.10801},
archivePrefix={arXiv},
primaryClass={cs.RO}
}

We present a hierarchical tree-based motion planning strategy, HAS-RRT, guided by the workspace skeleton to solve motion planning problems in robotics and computational biology. Relying on the information about the connectivity of the workspace and the ranking of available paths in the workspace, the strategy prioritizes paths indicated by the workspace guidance to find a valid motion plan for the moving object efficiently. In instances of suboptimal guidance, the strategy adapts its reliance on the guidance by hierarchically reverting to local exploration of the planning space. We offer an extensive comparative analysis against other tree-based planning strategies and demonstrate that HAS-RRT reliably and efficiently finds low-cost paths. In contrast to methods prone to inconsistent performance across different environments or reliance on specific parameters, HAS-RRT is robust to workspace variability.

@ARTICLE{9851528,
author={Uwacu, Diane and Yammanuru, Ananya and Morales, Marco and Amato, Nancy M.},
journal={IEEE Robotics and Automation Letters},
title={Hierarchical Planning With Annotated Skeleton Guidance},
year={2022},
volume={7},
number={4},
pages={11055-11061},
doi={10.1109/LRA.2022.3196885}}

We present a hierarchical skeleton-guided motion planning algorithm to guide mobile robots. A good skeleton maps the connectivity of the subspace of c-space containing significant degrees of freedom and is able to guide the planner to find the desired solutions fast. However, sometimes the skeleton does not closely represent the free c-space, which often misleads current skeleton-guided planners. The hierarchical skeleton-guided planning strategy gradually relaxes its reliance on the workspace skeleton as C-space is sampled, thereby incrementally returning a sub-optimal path, a feature that is not guaranteed in the standard skeleton-guided algorithm. Experimental comparisons to the standard skeleton guided planners and other lazy planning strategies show significant improvement in roadmap construction run time while maintaining path quality for multi-query problems in cluttered environments.