Date of Award

5-2021

Publication Type

Thesis

Degree Name

M.A.Sc.

Department

Electrical and Computer Engineering

Keywords

SAUV Balancing, Semi‐Autonomous Underwater Vehicle (SAUV), Vision‐based blob detection, Vision‐based tracking, Pipeline tracking, Leakage detection

Supervisor

S. Alirezaee

Supervisor

M. Saif

Rights

info:eu-repo/semantics/openAccess

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Abstract

This thesis intends to convert a Remote Operated Vehicle (ROV) to a Semi-Autonomous Underwater Vehicle (SAUV) using a vision-based control system. The SAUV was used for automatic underwater gas pipeline tracking and leakage detection. the leakages in the pipeline using Computer Vision. The SAUV was designed to operate both manually and automatically in underwater conditions. The proposed SAUV has 6 thrusters to achieve 4 degrees of freedom controlled by the controller unit and powered by LiPo battery packs. Our underwater vehicle is equipped with sensors providing continuous feedback signals to automatically control the vehicle to track predefined trajectories. The SAUV can be self-stabilized as the center of gravity and center of buoyancy of the vehicle is positioned in such a way in the predefined plan. The SAUV captures images to perform line tracking along with the pipeline and gas bubble images during its mission. The multi-core umbilical cable is used here for the video signal, the feedback signal, and battery charging lines. This will be used only for development and test purposes and will be removed during autonomous missions. For performing all operations, various control schemes such as computer vision algorithm for object detection using python programming, OpenCV, Hough Transform Theory, etc. are applied. The proposed SAUV is expected to pave the way for the development of advanced underwater oil and gas pipeline industrial applications by ocean scientists.

Share

COinS