forest_gen

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, [email protected]
Maintainers: Zachary Taylor, [email protected] and Helen Oleynkova, [email protected]
Affiliation: Autonomous Systems Lab, ETH Zurich

Bibliography

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.

[PDF here]

@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}
}

Installation Requirements

• ROS indigo and up
• rotors_simulator for octomap generation from Gazebo worlds.
• Recommended: volumetric_mapping for using Octomap as a ROS library within other programs.

Package Contents

• 10 pre-generated forest maps, 10 meters by 10 meters by 10 meters, with a density of 0.2 trees/square meter. (worlds)
• 1 pre-generated large map, 50 meters by 50 meters, with a density of 0.1 trees/square meter. (big_worlds)
• Royalty-free tree models, created by Miedevalworlds
• Generation script to create environments of any size and tree density.
• Start and end points used for planning evaluation in [1], as a CSV.

Planning Evaluations

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.

Generating New Environments

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