LES

LES implements Locally Exploitative Sampling, a framework that concentrates new samples where they can still improve the current solution rather than spreading them uniformly over the search space.

The repository is a ROS catkin package built on the Open Motion Planning Library (OMPL), and ships four planners: RRTsharp, conventional RRT# with informed sampling; ESTRELRRTsharp, RRT# with the relevant-region sampling framework; LESDRRT, a Deformable RRT variant; and TRUSTRRTsharp, RRT# running the LES framework itself. A Python script plots the resulting paths.

Reference

S. Joshi, S. Hutchinson and P. Tsiotras, “LES: Locally Exploitative Sampling for Robot Path Planning,” IEEE International Conference on Robotics and Automation, London, UK, May 29 – June 2, 2023, pp. 1551–1557. doi:10.1109/ICRA48891.2023.10160279 · PDF

Released under the GNU General Public License v3. See the Robust Autonomy for Uncertain Systems using Randomized Search project page for the wider context.

Visit the GitHub repository for the source code.