Multi-Sensor Fusion for Enhanced Navigation

Final Year Individual Project by Xinyang Huang, University College London. This project is designed and tested on Ubuntu 22.04, ROS2 Humble.

Table of Contents

• Introduction
• Project Report
  • Report in PDF
  • Original LaTeX
• Code
  • depth_image_to_laserscan
  • robot_urdf
  • multi_sensor_fusion
• Test Results
• Future Work

Introduction

This repository contains all the necessary code, documentation, and additional resources for the "Multi-Sensor Fusion for Enhanced Navigation" project. The project aims to improve robotic navigation capabilities by integrating data from multiple sensors to overcome the limitations of 2D LiDAR systems.

Project Report

Report in PDF

Find the comprehensive project report in PDF format here.

Original LaTeX

Access the LaTeX source files used to generate the project report here.

Code

This section includes all the source code developed for the project, organized into specific modules:

depthimage_to_laserscan

Code for converting depth images to laser scan data, enhancing obstacle detection in 2D navigation systems. More info

robot

Implementation of the robotic control algorithms including navigation and sensor integration. More info

multi_sensor_fusion

Core algorithms for the fusion of multiple sensor data aimed at providing accurate real-time localization and mapping. More info

Test Results

Documentation of testing procedures, results, and how they validate the effectiveness of the proposed solutions.

With 2D LiDAR only:

With the fusion algorithm:

Barricades world

With 2D LiDAR only:

With the fusion algorithm:

Future Work

Outline of potential future extensions and improvements to the project, based on current outcomes and technological advancements.

• Test on real robots.
• Use multiple rows in depth image rather than one row.
• Use multiple cameras to achieve 360 degree detection.

TODO

Add setup guidance for this project

---

For more information or inquiries, please contact Xinyang Huang.