ui_sim

This repository contains all the ROS packages necessary to run the Under-Ice UUV simulation in Gazebo, as described in [1].

Disclaimer: this is research-grade code! While the mathematical concepts employed are (hopefully!) sound, the implenetation may be a bit hacky. Additionally, this git repository contains multiple ROS packages, as well as the source files for GeographicLib (1.48). I acknowledge it's not a good practice to package multiple and unrelated discrete projects in a single repo, but it's the quickest way to allow others to get up and running.

Discrete ROS packages contained in this repository:

• uuv_simulator (Source: UUV Simulator
• hector_gazebo (Source: Hector Simulator
• laughlin_uw_plugins
• tdma_broadcaster
• under_ice_sim

Install External Deps

For UUV_Simulator meta package (per UUV_Simulator Wiki):

sudo apt-get install ros-kinetic-gazebo-msgs ros-kinetic-gazebo-plugins ros-kinetic-gazebo-ros ros-kinetic-gazebo-ros-control ros-kinetic-gazebo-ros-pkgs ros-kinetic-effort-controllers ros-kinetic-image-pipeline ros-kinetic-image-common ros-kinetic-perception ros-kinetic-perception-pcl ros-kinetic-robot-state-publisher ros-kinetic-ros-base ros-kinetic-viz python-wstool python-catkin-tools python-catkin-lint ros-kinetic-hector-localization ros-kinetic-joy ros-kinetic-joy-teleop libopencv-dev protobuf-compiler protobuf-c-compiler ros-kinetic-gps-common

For trajectory controller in UUV_Simulator (the trajectory tracker wants numpy and scipi, don't recall where/how I landed on installing other packages)

sudo apt-get install python-numpy python-scipy python-matplotlib ipython ipython-notebook python-pandas python-sympy python-nose

Install GeographicLib (I use this in my modified GPS plugin for better cartesian-WGS84 ellipsoid conversion):

cd ~/ros_ws/src/under_ice_meta
tar -zxvf GeographicLib-1.48.tar.gz
cd GeographicLib-1.48
mkdir build
cd build/
cmake ..
make
sudo make install

Build and run

Define a few paths

Prior to launching the simulation, you need to modify three paths defined in prepExpResultsFolder.sh. This script is run with ice.launch and creates a c onsistant folder structure, where bag files are stored for later use, as well as copies of all ice, vehicle, etc. xacro files. This can be useful for determining, "What ice velocity did I use in that experiment?..."

Suppose you wish to log data to /sim_data

(neglecting folder permissions ownership, yadda yadda)

mkdir /sim_data 

Now change the following three paths to reflect your desired save location, and full path to .../under_ice_sim/models/...

roscd under_ice_sim/scripts/
vim prepExpResultsFolder.sh

Change the following paths to reflect where you wish to save data, and location of the ice and vehicle xacro files:

save_path="/sim_data"
ice_model_path="/home/laughlin/ros_ws/src/under_ice_sim/models/ice"
veh_model_path="/home/laughlin/ros_ws/src/uuv_simulator/uuv_descriptions/models/rexrov/urdf"

Lastly, change the default path where rosbags are to be saved by editing data_record.launch to reflect the appropriate paths (as above):

roscd under_ice_sim/launch
vim data_record.launch

Change each of the three paths to read args="--output-name /sim_data/$(arg folder)/bags/measurements.bag ...

Prior to launching a simulation, create a folder within /sim_data to store this simulation's results:

mkdir /sim_data/simulation1

To launch the simulation, we will launch 6 distinct launch files. They should be launched in this order, so use Terminator, tmux, or screen to give yourself at least six prompts:

roslaunch under_ice_sim empty_underwater_world.launch timeout:=1000
roslaunch uuv_descriptions upload_rexrov_ui.launch x:=1050 y:=0 z:=-30 yaw:=3.14
roslaunch under_ice_sim ice.launch folder:=simulation1
roslaunch under_ice_sim vehicle_pid_controller.launch
roslaunch under_ice_sim send_waypoints_lawnmower_1000x100_6_10.launch
roslaunch under_ice_sim data_record.launch folder:=simulation1

Note:

1. By default Gazebo is run in a headless mode to reduce CPU load. RVIZ can be used for visualization. To change this, modify the appropriate line in empty_underwater_world.launch

If you wish to run the EKF you will need to do a few things as well.

Edit ui_ekf/launch/defaultEKF.launch to reflect:

<arg name="output_path" value="/sim_data/$(arg sim_folder)/$(arg output_folder)" />

Create and experiment folder where EKF results will be saved (I create a seperate folder for each "experiment" I do):

mkdir /sim_data/simulation1/ekf_test01

Copy the default EKF config file to the experiment folder (this contains all the parameters needed by the EKF)

roscd ui_ekf/config
cp default.yaml /sim_data/simulation1/ekf_test01

The EKF should now be ready to run:

roslaunch ui_ekf defaultEKF.launch sim_folder:=simulation1 output_folder:=ekf_test01

To unpause Gazebo from the command line:

gz world -p 0

Laughlin Barker, January 17, 2018

[1] Barker, L. D. L., and Whitcomb, L. L.,"A Preliminary study of ice-relative undwater vehicle navigation beneath moving sea ice," IEEE International Conference on Robotics and Automation (ICRA), 2018. Accepted.