Use of machine learning and robotic sensing to target renewals in concrete gravity sewers

Stephen Barclay Stephen, Michael Kacprzak, Boyu Li, Ting Guo, Yang Wang, Vinoth Viswanathan, Sarath Kodagoda, Karthick Thiyagarajan, Dammika Vitanage

Research output: Chapter in Book / Conference PaperConference Paper

Abstract

![CDATA[Sydney Water has approximately 823km of gravity concrete sewers where concrete corrosion is a widespread issue. Early identification of corrosion levels is important to make decisions on fit-for-purpose renewal methods. Sydney Water and the University of Technology Sydney (UTS) have collaborated in developing machine learning models and robotic systems to identify corrosion levels in the carriers as well as to quantitatively measure the depth of corrosion and the depth of reinforcement bars. The machine learning model predicts corrosion hotspots in the carrier, which are further inspected by a manhole deployable robotic system. The new robotic observations are fed back into the machine learning model for continuous improvement. This integrated system has many advantages compared to the current practice.]]
Original languageEnglish
Title of host publicationThe Australian Water Association Annual Conference, OzWater'23, 9-12 May 2023
PublisherThe Australian Water Association
Number of pages7
Publication statusPublished - 2023
EventOzWater (Conference) -
Duration: 1 Jan 2023 → …

Conference

ConferenceOzWater (Conference)
Period1/01/23 → …

Fingerprint

Dive into the research topics of 'Use of machine learning and robotic sensing to target renewals in concrete gravity sewers'. Together they form a unique fingerprint.

Cite this