GADEN is a simulation framework designed for mobile robotics systems and gas sensing algorithms, also known as Mobile Robotics Olfaction (MRO). The framework is rooted in the principles of computational fluid dynamics and filament dispersion theory, modeling wind flow and gas dispersion in 3D scenarios and accounting for walls, furniture, and other objects that may have a significative impact on the gas dispersion.
Moreover, it integrates the simulation of different environmental sensors, such as metal oxide gas sensors, photo-ionization detectors, T-DLAS or anemometers, as well as it is fully integrated with ROS and the navigation stack, making testing and validation much easier.
A demonstration video can be seen at the MAPIRlab channel on YouTube.
You can cite Gaden in your work using the following .bib:
@article{monroyGADEN3DGas2017,
title = {{{GADEN}}: {{A 3D}} Gas Dispersion Simulator for Mobile Robot Olfaction in Realistic Environments},
author = {Monroy, Javier and {Hernandez-Bennetts}, Victor and Fan, Han and Lilienthal, Achim and {Gonzalez-Jimenez}, Javier},
year = {2017},
journal = {Sensors (Switzerland)},
volume = {17},
number = {7},
pages = {1--16},
doi = {10.3390/s17071479},
}
Gaden 2 (this branch) supports only Humble, for now. You will likely run into compilation issues when using different versions. When a new ROS LTS is released we will probably migrate the project to it.
Move to your colcon workspace and run
git clone --recurse-submodules [email protected]:MAPIRlab/gaden.git src/gaden
It is important that you include the --recurse-submodules
option, as the gaden repository includes submodules. if you already cloned in a non-recursive manner you can fix the problem by running
git submodule update --init --recursive
Although GADEN is a self-contained pkg, the also included "simulated_sensor_pkgs" (anemometer and gas sensors) depends on an external pkg defining some "olfaction" related msgs. This pkg is available in a different repository olfaction_msgs
See the tutorial for instructions on how to set up a simulation of your own or use the included test environments.
Users who want to test their algorithms in complex environments but do not need to simulate any specific scenario can find an extensive repository of existing simulations in the VGR dataset, which features simulations in 3D models of real houses, along with the configuration files and intermediate data.