Planning navigation for SIAR robotics platform.
siar_navigation allows you to implement a navigation system based in a controller, costmap and a planner. The controller is implemented in the siar_controller package and allows avoid positive and negative obstacles. The costmap is implemented in siar_costmap package and detects positive and negative obstacles from sensor data and provides the user with a cost map taking into account the position of the obstacles. The planner is implemented in the siar_planner package and computes a path taking into account a costmap. The planner can use the next algorithms to compute the path: RRT, bi-RRT, t-RRT and t-bi-RRT.
We recommend to use the provided docker for testing the developments.
Just build the provided DockerFile in its folder:
$ cd docker
$ docker build . -t siar
Then, you can start the docker with the provided file and run the experiments as listed in section Experiments
$ ./run_container_linux.bash
To compile and execute successfully siar_navigation is necessary to install compile the next packages into a Ubuntu 16.04 system with ROS Kinetic:
- costmap-2d, navigation, gazebo-ros, gazebo-ros-pkgs, gazebo-ros-control and libann-dev.
sudo apt-get install ros-kinetic-costmap-2d
sudo apt-get install ros-kinetic-navigation
sudo apt-get install ros-kinetic-gazebo-ros
sudo apt-get install ros-kinetic-gazebo-ros-pkgs
sudo apt-get install ros-kinetic-gazebo-ros-control
sudo apt-get install libann-dev
From the repository robotics-upo are necessary the next packages:
- function [https://github.com/robotics-upo/functions.git]
- siar_package (banch kinetic) [https://github.com/robotics-upo/siar_packages.git]
- arduimu_v3 [https://github.com/robotics-upo/arduimu_v3.git]
- depth2cloud [https://github.com/robotics-upo/depth2cloud.git]
Here you can find the instruction for executing the simulations presented in the paper "Advanced sampling-based planning strategies for a sewer inspection reconfigurable robot" submitted to the Robotics journal of MDPI. The three different planners were tested in the in a testbench of simulations in synthetic maps, where the planner with the best performance was t-RRT. Finally, the RRT* planner was tested in realistic simulations using the Gazebo high fidelity robot simulator. The model and world used in the simulations can be found at [https://github.com/robotics-upo/siar_simulator].
All the code was developed and tested in Ubuntu 18.04 and ROS Melodic.
Run the following command to execute all the synthetic map simulations presented in the paper:
rosrun siar_planner test_all_algorithms_synthetic.sh <number_of_tests>
It allows you to execute all the proposed tests in the scenarios a given number of times. E.g, for ten times:
rosrun siar_planner test_all_algorithms.sh 10
To execute the Gazebo simulations, it is necessary to follow the next steps:
-
Download the siar_simulator package into a ROS workspace from [https://github.com/robotics-upo/siar_simulator]
-
Compile the ROS workspace
catkin_make
- Execute:
roslaunch siar_gazebo siar_simulator_complete_T130_gut30.launch
to have a SIAR with seven cameras, or:
roslaunch siar_gazebo siar_simulator_complete_T130_gut30_velodyne.launch
to have SIAR with six cameras and one velodyne.
IMPORTANT: this launchers will start the simulator in pause mode to avoid conflict when spawing the different models. The siar_costmap module will generate some errors that will no longer thrown just by pushing play in the Gazebo simulator.
- Execute:
roslaunch siar_planner planner_action_server_simulation.launch
To use differents planners just change the parameter planner_type from to one of the following options: 'rrt', 'trrt', 'birrt', 'tbirrt'.
Congratulations !!! now you are ready to navigate with SIAR in sewer environment.