BlueROV2 MPC

This work implements a NMPC controller for BlueROV2 with ROS.

This repository contains the robot description and necessary launch files to simulate the BlueROV2 (unmanned underwater vehicle) on Unmanned Underwater Vehicle Simulator (UUV Simulator). Additional it's possible run BlueROV2 in SITL using mavros, joystick interaction and video streaming capture with opencv based on bluerov_ros_playground package from BlueRobotics.

This work is developed based on Ingeniarius, Lda. and Instituite of Systems and Robotics University of Coimbra within the scope of MS thesis "Localization of an unmanned underwater vehicle using multiple water surface robots, multilateration, and sensor data fusion".

Prerequisites

• Python 3.7
• ROS (ROS noetic recommended)
• uuv simulator
• MAVROS
• Acados

We also provide its docker image to save you some time.

Getting started

1. Install python 3.7 and its dependencies

   sudo add-apt-repository ppa:deadsnakes/ppa
   sudo apt update
   sudo apt install -y python3.7
   
   apt-get install -y python3-pip
   
   pip3 install numpy matplotlib scipy future-fstrings casadi>=3.5.1 setuptools
   sudo apt-get install -y python3.7-tk
   

2. Install Acados

   git clone https://github.com/acados/acados.git
   cd acados
   git submodule update --recursive --init
   mkdir -p build
   cd build
   cmake -DACADOS_WITH_QPOASES=ON -DACADOS_WITH_OSQP=OFF/ON -DACADOS_INSTALL_DIR=<path_to_acados_installation_folder> ..
   make install -j4
   
   pip install -e ~/acados/interfaces/acados_template
   
   echo 'export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:"/root/acados/lib"' >> ~/.bashrc 
   echo 'export ACADOS_SOURCE_DIR="/root/acados"' >> ~/.bashrc
   source ~/.bashrc
   

3. Install ros uuv-simulator packages

   sudo apt-get install -y ros-noetic-geodesy
   mkdir -p ~/catkin_ws/src && \
   cd ~/catkin_ws/src && \
   git clone --branch noetic https://github.com/arturmiller/uuv_simulator.git
   

4. Install then compile this package

   cd ~/catkin_ws/src && \
   git clone https://github.com/HKPolyU-UAV/bluerov2.git
   

   Please edit CMakelists.txt here and here if you put your acados in otherwise directory. Modify the line set(acados_include "~/acados/include") and set(acados_lib "~/acados/lib")

   Then do

   cd ~/catkin_ws/src/bluerov2/bluerov2_dobmpc/scripts && \
   yes | python3 generate_c_code.py
   

   and

   cd ~/catkin_ws/src/bluerov2/bluerov2_mpc/scripts && \
   yes | python3 generate_c_code.py
   

   The last commands basically run generate_c_code.py at here and here.

   Finally, do

   cd ~/catkin_ws && catkin_make
   

Start simulation without controller

Ocean_waves in uuv simulator is set as default world.

Quick start to initialize the Gazebo world and add blueROV2 vehicle

roslaunch bluerov2_gazebo quick_start.launch

Or start with thruster manager

roslaunch bluerov2_gazebo start_with_thruster_manager.launch

With thruster manager, forces and moments can be published to topic /bluerov2/thruster_manager/input.

Start simulation with PID controller

Start PID controller with teleop (using joystick)

roslaunch bluerov2_gazebo start_pid_demo_with_teleop.launch

Or start PID Controller without teleop

roslaunch bluerov2_gazebo start_pid_demo.launch

Further trajectory tracking tasks can be done with uuv_control_utils package, such as following linear trajectory

roslaunch uuv_control_utils send_waypoints_file.launch uuv_name:=bluerov2 interpolator:=linear

Or following helical trajectory

roslaunch uuv_control_utils start_helical_trajectory.launch uuv_name:=bluerov2 n_turns:=2

Start simulation with MPC controller

A demonstration of starting MPC controler to follow circular trajectory can be launched by

roslaunch bluerov2_gazebo start_mpc_demo.launch

The bluerov2_mpc package also includes more trajectory following tasks such as following leminscate path and self defined waypoints.

Start simulation with disturbance observer-based MPC controller

A demonstration of starting MPC controler to follow circular trajectory can be launched by

roslaunch bluerov2_dobmpc start_dobmpc_demo.launch 

The generated disturbances can be adjusted at here or here.

Citation

This repository contains the research code for our paper: https://doi.org/10.3390/jmse12010094.