The repo aims to provide a planning algorithm and relevant packages to enable a saliency--aware autonomous exploration task for aerial robots.
Create a new catkin workspace and clone all modules into that workspace.
mkdir vsep_ws && cd vsep_ws && mkdir src && cd src
git clone https://github.com/unr-arl/vseplanner
git submodule update --init --recursive
cd ..
catkin config -DCMAKE_BUILD_TYPE=Release
catkin build
We tested this repo with ROS Jade on Ubuntu 14.04, update for newer versions like ROS Kinetic will come soon.
First, unzip the gazebo_models.zip in the planner folder into the ./gazebo folder:
unzip gazebo_models.zip && mv gallery_plane gallery_wall painting_wall ~/.gazebo/models
Then, launch the simulation file:
roslaunch ./src/launch/sim_vsep.launch
We provided here two datasets recored onboard of a MAV. (More details are explained here). To test the planner, please run the rosbags as well as launching a file:
rosbag play VSEP_Arena.bag --clock
roslaunch ./src/launch/mav_vsep_arena_offline.launch
and use a ros service command shown as below to trigger the planner.
rosservice call /vseplanner '{header: {stamp: now, frame_id: world}}'
If you use this software in a scientific publication, please cite the following paper:
@inproceedings{tung2018vsep,
title={Visual Saliency-aware Receding Horizon Autonomous Exploration with Application to Aerial Robotics},
author={Dang, Tung and Papachristos, Christos and Alexis, Kostas},
booktitle={2018 IEEE International Conference on Robotics and Automation (ICRA)},
pages={to appear},
year={2018},
organization={IEEE}
}
You can contact us for any question or remark: