Pipeline Leak Identification Emergency Robot Swarm (PLIERS)

Loading...
Thumbnail Image

Authors

Kandil, Ayman
Badran, Habib
Hassan, Osama
Bin Jassem, Maryam

Issue Date

2021-01-31

Type

Language

Keywords

ELEG/CPEG

Research Projects

Organizational Units

Journal Issue

Alternative Title

Abstract

In our continuously expanding cities, the demand for industrial factories, sewage systems, and power stations increases exponentially. Maintaining these huge factories and stations autonomously is becoming increasingly important. Pipelines are especially hard to investigate and examine by humans; therefore, automated robots play an important role in identifying any cracks or damage. In this project, we introduce the Pipeline Leak Identification Emergency Robot Swarm (PLIERS), which aims to detect cracks in pipes and notify the operators whenever a pipe needs to be replaced. In PLIERS, swarm robots investigate and scan the pipes from the inside for cracks and save the locations of detected cracks on a map of pipes. The swarm robots depend on the ZigBee standard to communicate among themselves. The ZigBee standard is a stack of communication protocols that is well-suited for short-range data communication. ZigBee is well-known for its power efficiency and flexibility. The swarm robots are equipped with ultrasonic sensors to detect any cracks or damages in pipes. Then, cameras collect images about the inside of the pipes and feed them into a module that implements machine learning (ML) algorithms. These algorithms are used to detect or even expect the occurrence of cracks or damages. Furthermore, ML algorithms are used to identify the crack's severity by applying an intensive analysis of the captured images. The swarm robots send the damaged pipe's location and damage severity to a control station to notify the maintenance team. PLIERS is designed to provide an efficient and safe way of maintaining pipelines without any human intervention.

Description

Citation

Publisher

College of Engineering and Applied Sciences

License

Journal

Volume

Issue

PubMed ID

DOI

ISSN

EISSN

Collections