Robotic Fish Enabled Offshore Pipeline Inspection

Abstract

Abstract Timely inspection of subsea infrastructure, especially subsea pipelines, is the key to the prevention of oil spills. In this paper, a transformative offshore pipeline inspection technology is presented by using a bio-inspired autonomous robotic system equipped with a processing unit for underwater computer vision processing and edge computing. The goal is to build a time-efficient and cost-effective system for underwater pipeline inspection that can detect oil leakage at early stages and prevent disastrous results. In this paper, we introduced a bio-inspired autonomous underwater vehicle (BAUV) equipped with video cameras and mobile edge computing devices. We deploy a deep neural network (DNN) specially trained for a variety of underwater image/video processing tasks. The intelligent computer vision processing unit allows us to navigate and track objects even when the visibility is poor. This time-efficient and cost-effective solution will detect pipeline leakage and rupture at an early stage and allow operators to make timely and informed decisions to minimize environmental impacts.