Generates randomized Poisson forests to use for UAV collision avoidance evaluations.
Input: map size and density Output: Gazebo world file and (optionally) Octomap
Authors: Zachary Taylor, zachary.taylor@mavt.ethz.ch \ Maintainers: Zachary Taylor, zachary.taylor@mavt.ethz.ch and Helen Oleynkova, helen.oleynikova@mavt.ethz.ch \ Affiliation: Autonomous Systems Lab, ETH Zurich
If using these datasets, please cite [1]:
Helen Oleynikova, Michael Burri, Zachary Taylor, Juan Nieto, Roland Siegwart, and Enric Galceran, “Continuous-Time Trajectory Optimization for Online UAV Replanning”. In IEEE Int. Conf. on Intelligent Robots and Systems (IROS), 2016.
@inproceedings{oleynikova2016continuous-time,
author={Oleynikova, Helen and Burri, Michael and Taylor, Zachary and Nieto, Juan and Siegwart, Roland and Galceran, Enric},
booktitle={IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
title={Continuous-Time Trajectory Optimization for Online UAV Replanning},
year={2016}
}
start_and_end.csv
contains 900 free start and end points, used for the planning evaluations in [1]. All points are at least 4 meters apart and both start and goal are free space, given a bounding box of 1.2 meters by 1.2 meters by 1.0 meters. The actual bounding box used for path evaluations was 0.2 meters smaller; 1.0 meters by 1.0 meters by 0.8 meters.
Use the following shell script to generate new worlds.
./genForests.sh <number of worlds to gen> <world side length> <tree density> <octomap res>
./genForests.sh 1 10 0.2 0.2