Skip to content
BranchesTags

Latest commit

 

History

27 Commits

Folders and files

NameName
Last commit message
Last commit date
models
models
 
 
src
src
 
 
vrx_world
vrx_world
 
 
.gitattributes
.gitattributes
 
 
.gitignore
.gitignore
 
 
LICENSE
LICENSE
 
 
README.md
README.md
 
 

Repository files navigation

🚤 Autonomous ASV Docking and Path Tracking

MATLAB / Simulink + ROS 2 + Gazebo (VRX)

ROS 2 MATLAB Simulink Gazebo Ubuntu VRX

A complete Guidance, Navigation, and Control (GNC) framework for a WAM‑V Catamaran Autonomous Surface Vessel (ASV) validated in the VRX (Virtual RobotX) high‑fidelity simulation environment.

This project integrates:

  • 3‑DOF vessel dynamics modeling
  • Adaptive Circular Line‑of‑Sight (LOS) guidance
  • PD‑based heading control
  • Surge velocity regulation
  • Differential thrust mixing (twin thrusters)
  • Real‑time ROS 2 communication
  • Precision autonomous docking strategy

✅ Smooth path tracking and zero‑overshoot docking without RPM discontinuities.

🎥 Simulation Demonstration (2 min)

This demonstration showcases:

  • Adaptive Circular LOS path tracking
  • Smooth differential thrust control
  • Turn-then-go heading stabilization
  • Precision zero-overshoot docking

Autonomous ASV Docking Demo

▶ Click the image above to watch the full VRX simulation on YouTube.

📌 Table of Contents

✨ Key Features

  • Adaptive Circular LOS Guidance
    Continuous path progression without waypoint switching spikes.

  • Smooth Differential Thrust Profile
    Eliminates saw‑tooth RPM oscillations caused by discrete waypoint jumps.

  • Distance‑Based Docking Deceleration
    Square‑root velocity shaping for precise, zero‑overshoot stopping.

  • Turn‑Then‑Go Strategy
    Reduces surge speed during large heading errors for safe maneuvering.

  • Real‑Time Simulink ↔ ROS 2 Integration
    Live GPS and IMU feedback from Gazebo with real‑time thrust publishing.

🧠 System Architecture

flowchart LR
  P["Path / Markers"] --> G["Adaptive Circular LOS"]
  G --> H["PD Heading Control"]
  H --> S["Surge Speed Controller"]
  S --> M["Differential Thrust Mixer"]
  M --> V["WAM-V Dynamics<br/>(VRX Gazebo / VRX)"]
  V --> F["GPS + IMU Feedback"]
  F --> G
Loading

Control Design Highlights

  • Continuous spline‑based path tracking
  • Adaptive lookahead radius near docking zone
  • Heading latch for final approach stability
  • Differential thrust control for twin‑hull propulsion

🛠 System Requirements

🖥 Operating System

📦 MATLAB / Simulink Requirements

This project was developed and validated using:

  • MATLAB R2025b Update 3
  • Simulink (R2025b)

🔧 Required Toolboxes

All toolboxes below are actively used within the Simulink model and required for full functionality:

  • Aerospace Blockset
  • Aerospace Toolbox
  • Automated Driving Toolbox
  • Computer Vision Toolbox
  • Control System Toolbox
  • Image Processing Toolbox
  • ROS Toolbox (ROS 2 interface)
  • Simulink Control Design

⚠️ Missing toolboxes may cause unresolved block errors or simulation failures.

🔍 Verify Installed Toolboxes

Run inside MATLAB:

ver

Ensure all required toolboxes appear before running the model.

📂 Repository Layout

Your modified VRX world file is version‑controlled inside this repository at:

vrx_world/sydney_regatta.sdf

Recommended structure:

asv-autonomous-docking-and-path-tracking/
├── src/
│   ├── functions/
│   └── main/
├── scripts/
│   └── update_path_sydney_regatta.py
├── vrx_world/
│   └── sydney_regatta.sdf          ✅ (Modified VRX world file)
├── docs/
└── results/

🚀 Installation & Setup

1️⃣ Install VRX (Mandatory)

Follow official instructions:

https://github.com/osrf/vrx

2️⃣ Clone This Repository

git clone https://github.com/kavindagehan/asv-autonomous-docking-and-path-tracking.git
cd asv-autonomous-docking-and-path-tracking

3️⃣ Source ROS 2 (and VRX workspace if applicable)

source /opt/ros/jazzy/setup.bash

If VRX was built in a workspace:

source ~/vrx_ws/install/setup.bash

4️⃣ Replace VRX sydney_regatta.sdf (REQUIRED — do this BEFORE MATLAB)

This project depends on a modified sydney_regatta.sdf that contains the marker tags:

<!-- START-MARKERS -->
<!-- END-MARKERS -->

These tags are required for automatic path injection (triggered by the MATLAB live script).

4.1 Locate your VRX worlds directory

Case A — VRX built from source (common):

VRX_WORLDS_DIR=~/vrx_ws/src/vrx/vrx_gz/worlds

Case B — VRX available via package share directory:

VRX_WORLDS_DIR="$(ros2 pkg prefix vrx_gz)/share/vrx_gz/worlds"

Verify the directory exists:

ls "$VRX_WORLDS_DIR"

4.2 Backup the original world file

cd "$VRX_WORLDS_DIR"
mv sydney_regatta.sdf sydney_regatta_original.sdf

4.3 Copy the modified world file from this repository

Run this from the repo root:

cp vrx_world/sydney_regatta.sdf "$VRX_WORLDS_DIR/"

4.4 Quick validation (confirm marker tags exist)

grep -n "START-MARKERS\|END-MARKERS" "$VRX_WORLDS_DIR/sydney_regatta.sdf"

✅ If both markers are printed, the world file is correctly installed.

▶️ Execution Order

Path injection is performed automatically when running: ASV_SIM_ROS_MAIN_V2_1.mlx.

1) Open MATLAB / Simulink

Start MATLAB first (recommended), then run the main live script:

  • Run: ASV_SIM_ROS_MAIN_V2_1.mlx
    • This automatically performs the required path injection into the VRX world file
    • It also configures ROS 2 communication required by the Simulink model

2) Build VRX workspace (if VRX is built from source)

cd ~/vrx_ws
colcon build --merge-install

If VRX is already built and unchanged, you can skip this build step.

3) Launch the VRX simulation

ros2 launch vrx_gz competition.launch.py world:=sydney_regatta

4) Run the Simulink model

Open and run:

  • Simulink model: Simulink_ROS_Boat_Path_Follow_V2_5.slx

(Ensure the model is in Run mode and publishing thrust commands.)

5) Verify ROS topics (optional check)

List topics:

ros2 topic list

Check thrust output:

ros2 topic echo /wamv/thrusters/left/thrust

🔁 ROS Interface

Subscribed Topics

  • /wamv/sensors/gps/gps/fix
  • /wamv/sensors/imu/imu/data

Published Topics

  • /wamv/thrusters/left/thrust
  • /wamv/thrusters/right/thrust

📌 Notes

  • Environmental disturbances (wind, waves, current) can be enabled in VRX for robustness testing.
  • Validation was performed under still‑water conditions for baseline benchmarking.
  • The original world file is preserved as a backup (sydney_regatta_original.sdf).
  • If the VRX world file does not contain the marker tags, the MATLAB live script’s automatic path injection will fail.

👨‍🎓 Author

Gehan Kavinda Dasanayake
Research Project – Carinthia University of Applied Sciences

📜 License

This project is licensed under the MIT License. See the LICENSE file for details.

About

A complete Guidance, Navigation, and Control (GNC) framework for WAM-V Autonomous Surface Vessels using MATLAB/Simulink, ROS 2, and Gazebo (VRX) for adaptive path tracking and autonomous docking.

Topics

asvautonomous-dockingautonomous-surface-vesselcontrol-systemsgazebogncmarine-roboticsmatlabpath-trackingros2-jazzysimulinkusvvrxwam-v

Resources

Readme
MIT license
Activity

Stars

23 stars

Watchers

1 watching

Forks

3 forks
Report repository

Releases

Packages

Contributors

Languages