ROS and Gazebo based destruction scenarios for search and rescue robots.
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
destruction_scenarios
LICENSE.md
README.md

README.md

destruction_scenarios

This repository contains the source code for the Destruction Scenarios ROS package. The package provides two different destruction scenarios: a house/building and a garage. Each of these scenarios are available in three forms: easy, intermediate and hard. These "difficulty levels" were chosen according to the level of destruction, i.e., according to the difficulty of traversing the terrain on each level.

These levels have a launch file associated to each one, and use Gazebo to provide the simulation. According to the needs of each user, these launch files can be included in another, user-specified launch file that also launches the robot.

Our goal is to provide a simulation platform where Search and Rescue robot developers can test their robots in common destruction scenarios. We invite anyone to contribute to this package, and we have also made available a different repository with the scripts which were created to generate the scenarios that can be used to generate different ones (this second repository is available at https://github.com/roboptics/destruction_scenarios_generator).

Use Instructions

The package can be used just by launching one of the launch files, just as with any other ROS package. The only requirement is to point Gazebo to where the house and garage models are. To do this, the environment variable GAZEBO_MODEL_PATH must be set to the "worlds" folder of this package. Example: GAZEBO_MODEL_PATH=/path/to/catkin_ws/src/destruction_scenarios/worlds

License and Citing

The code and assets present in this repository are subject to the GPL license, as described in LICENSE.md.

If you wish to use the scenarios in your work and publish the results you obtain, please cite this work as such: "Destruction Scenario Dataset", http://www.roboptics.pt