Motion Planning Package

Description

This package deals with finding an optimal or sub-optimal route between an origin point and a termination point in marine environments. The underwater environment is still considered as a great challenge for the path planning of autonomous underwater vehicles (AUVs) because of its dynamic nature, So this package aims to make the AUV more adaptable to these challenges.

It consists of four modules (localization, global planner, local planner and motion control)

• Localization: determine AUV’s place in the 3D space around it and also determines the location of the obstacles and objects surrounding it.

• The Global Planner: give the AUV the ability to plan the correct and feasible path through the starting point to its goal.

• The Local Planner: generate a set of control commands that moves the AUV on the path generated by the Global Planner but by taking the dynamics into consideration while also trying to avoid obstacles.

• Motion Control: translate the set of control commands from the Local Planner into actions for the thrusters

License

The source code is released under a GNU GENERAL PUBLIC LICENSE

Author: vortex-co
Affiliation: VorteX-Co
Maintainer: vortex-co, info@vortex-co.com

The motion planning module package has been tested under ROS2 eloquent on Ubuntu 18.04

Table of contents

• Usage
• Config files
• Launch files
• Nodes
• Hardware

Usage

• Source your ROS2 installation

  source /opt/ros/eloquent/setup.bash
  

• Navigate to your workspace

  cd ~/$HOME/VAUV/Software/vortex_ws
  

• Build your package

  colcon build --packages-select <package_name>
  

• Start to run node by node with

  ros2 run motion_planning <node_name>
  

• or launch file with

  ros2 launch motion_planning launch_file_name.launch.py
  

Config files

…

Launch files

…

Nodes

pose_estimation node

Estimates 3D state of the AUV with the usage of DVL, IMU, point cloud data and visual odometry from the stereo camera.

Subscribed Topics

• /fused_date

Published Topics

• /estimated_pose

Services

..

Parameters

….

local_SLAM node

Generates a series of poses and sub-maps by matching this generated poses with a reference scan.

Subscribed Topics

• /estimated_pose

Published Topics

• /submaps

  Services

…

Parameters

…

global_SlAM node

Updates the poses/states generated from local_SLAM node and then updates the sub-maps to create a Global 3D map.

Subscribed Topics

• /submaps

  Published Topics

• /global_3D_map

Services

…

Parameters

…

searching node

It aims to find the optimal path from the AUV current location to it is goal and continuously validating this path.

Subscribed Topics

…

Published Topics

• /initial_path

Services

…

Parameters

…

path_smoothing node

Draws curved turns instead of sharp turns‏ by using a smoothing algorithm and the AUV’s local pose.

Subscribed Topics

• /initial_path

  Published Topics

• /smoothed_path

Services

….

Parameters

…

obstacle_detection node

Responsible for the process in which the upcoming objects in the path are detected using a deep learning model based on fused data from Sensors package.

Subscribed Topics

….

Published Topics

..

subnode_processing node

Regenerate a new path based on surrounding environment if an obstacle is detected.

Subscribed Topics

…

Published Topics

…

vector_processing node

Dealing with regenerated path from subnode_processing node and give the required motion profile for the control system of the AUV.

Subscribed Topics

…

Published Topics

…

motion_control node

Translate the set of control commands from the Local Planner into actions for the thrusters

Subscribed Topics

..

Published Topics

..

Services

…

Parameters

…

.

Hardware

This package accesses the following hardware:

• Pixhawk4