/Multigoal-Orchard-Drone-Planning-Library

A library for multigoal drone motion planning in orchard-like environments

Primary LanguageJupyter Notebook

Multigoal Agricultural Drone Planning Library (mgodpl)

Description

Multigoal Drone Planning Library (mgodpl) is a project developed to facilitate multi-goal drone planning in orchard-like environments.

Publications

W. Kroneman, J. Valente and A. F. Van Der Stappen, "A fast two-stage approach for multi-goal path planning in a fruit tree," 2023 IEEE International Conference on Robotics and Automation (ICRA), London, United Kingdom, 2023, pp. 1586-1593, doi: 10.1109/ICRA48891.2023.10160281. (https://ieeexplore.ieee.org/document/10160281)

File structure:

  • src/: Source code
    • src/math/: General-purpose math functions.
    • src/planning/: Path-planning algorithms and related utilities.
    • src/visualization/: Visualization utilities.
    • src/visualizations/: Various vizualizations, primarily focused on presentations and interaction.
    • src/experiments/: Code for running experiments with the goal of gathering statistics on algorithm performance.
    • src/experiment_utils/: Utilities relating to experiments and visualization without being used directly by the algorithms. (Only experiments and vizualizations should depend on this.)
  • src_old/: Old source code, kept for reference, but became unmaintainable.
  • test/: Unit tests of the code in src/.
  • test_robots/: Contains 3D models and robot descriptions for use in the experiments and vizualizations.
  • analysis/: Code for analyzing the results of experiments.