Skip to content
forked from MAPIRlab/gaden

GADEN: A 3D Gas Dispersion Simulator for Mobile Robot Olfaction in Realistic Environments

License

Notifications You must be signed in to change notification settings

aravindsrj/gaden

 
 

Repository files navigation

A 3D Gas Dispersion Simulator for Mobile Robot Olfaction in Realistic Environments

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, or anemometers, as well as it is fully integrated with ROS and the navigation stack, making testing and validation much easier. For further information and technical specifications, please see the project webpage at http://mapir.isa.uma.es/work/gaden.

TEST_ENV

Along with GADEN we also include in this repo a set of scenarios to help researchers to test and validate GADEN as well as their own algorithms (Gas Distribution Mapping, Gas Source Localization, etc.) in an easy way. To that end, the folder test_env contains multiple scenarios with pre-configured CAD models, wind-flow simulations and ROS-launch files that enable the user to easily start testing GADEN as well as their robotics solutions. Each scenario follows a similar structure:

  • cad_models: This folder contains all the cad models that define the environment. For convenience, we include not only walls and objects, but also the "inner" volume, that is, the free space region where gas is released (useful for CFD).

  • launch: This folder includes the launch and configuration files for testing GADEN and a simple Robotics simulation environment using Stage. Concretely, the files included are:

    1. GADEN_preprocessing: A launch file in charge of loading the CAD models (.stl ASCII) and CFD wind data to generate the Occupancy3D grid map and format the wind flows to the cubic-grid format used in GADEN. This must be the first launch file to run as it is a pre-processing stage necessary for the simulation.

    2. GADEN: This launch file is the one responsible for running the gas dispersal simulation, saving the results to the designated folder (see params). This script represents the core of GADEN.

    3. GADEN_player: This launch file is in charge of "playing back" a simulation by just reading the results from the previous phase and also providing different simulated gas/wind sensors. Its main utility is to visualize (RViz) the gas dispersal as a point cloud and to make the gas and wind data available in the ROS architecture to other packages.

    4. main_simbot: To illustrate the integration with ROS and move_base, this launch file set up the scenario, plays back the gas dispersal and introduces a mobile robot (equipped with different sensors) able to autonomous navigate the environment. This script can be used as a starting point for testing advanced MRO algorithms.

  • wind_simulations: All the scenarios include at least the results of one CFD simulation (see GADEN_tutorial.pdf) to enable the user to directly start testing the scenario without having to worry about CFD. Notice that the repository only includes the "wind_at_cell_centers.csv" files, that is, the results of CFD after exporting the wind vectors to a CSV file. It is then necessary to run "GADEN_preprocessing.launch" to re-format this raw data to the format required by GADEN.

olfaction_msgs

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

About

GADEN: A 3D Gas Dispersion Simulator for Mobile Robot Olfaction in Realistic Environments

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • C++ 96.8%
  • CMake 3.2%