Digital Twins and Robot Dogs: How AI Is Revolutionizing Infrastructure Upkeep

A critical part of modern asset management is understanding the asset’s condition and performance.

Does it need to be replaced or repaired, and is it performing optimally?

In time, human inspectors out in the field would have made these assessments, the classic worker in hardhat and clipboard ticking off items on a list.

However, in 2025, we will face a different world and a different future. This is one where sensors and robots can inspect assets and report in real time, with the data integrated into a digital twin.

In the U.K., for example, Boston Dynamics’s “dog” Spot robot is out in the field surveying sections of a motorway in Somerset.

At the site on the M5 motorway, Spot was scanning and capturing data on embankment sections using cameras and a Leica lidar tracker.

New Australian hub

Robotic inspection is also the rationale behind creating a new hub at the University of Sydney, launched to develop the use of robots and autonomous systems in collecting information on the state and performance of physical assets.

The ARC Australian Robotic Inspection and Asset Management Hub (ARIAM) was created in partnership with 15 industry partners, including Thales, Reach Robotics, Nexxis and Nearmap. Its aim is to equip robots to autonomously collect data that can be used to create digital twins.

During its five-year lifespan, the ARIAM Hub is expected to mobilize a team of more than 50 researchers and partner investigators to overcome the key technology barriers to the distribution of autonomous systems for asset management in real-time.

“We can eliminate the risk of injury associated with manned operations to inspect remote or dangerous areas.”

With suitable sensors, a new generation of robots can capture various aspects of a physical asset, such as its structural integrity, temperature, movement and other relevant symptoms. This data is integrated into the digital twin in real-time.

ARIAM’s work will reduce the need for people to enter dangerous or hazardous locations to maintain assets such as tunnels and underwater infrastructure. It will also help manage the looming ‘infrastructure cliff,’ which will see many post-World War II infrastructure assets approach their end of life.

Its work will lead to accurate and timely data collection, enhance safety, minimize disruption during maintenance and improve industry efficiency.

“Spider-like” robot

Professor Ian Manchester, from the School of Aerospace, Mechanical and Mechatronic Engineering, leads ARIAM researchers and directs the Australian Centre for Robotics.

The team plans to develop semi-autonomous and autonomous machines capable of inspecting and maintaining complex structures in challenging environments, removing the need for people to enter dangerous or hazardous locations.

In one example, Nexxis is developing a spider-like robot with magnetic feet capable of crawling around metal structures and inspecting them for damage.

The company’s mantra can be applied to ARIAM’s wider mission: “Nexxis is the result of a paradigm shift. This is where pioneering possibilities come to life.”

ARIAM argues that robotics’ role in asset management has conventionally been passive, chiefly involving repeated observation of assets and environments.

One of the hub’s goals is to advance active and interactive sensing and perception programs, progressing through stages of sophistication so that robots can “adaptively observe their environments, employ active perception techniques in which sensor reconfiguration forms part of the planning process.”

“This initiative marks a significant leap forward for robotics and will change the way industry operates,” Professor Manchester said.

“Infrastructure cliff”

“With new designs, we can eliminate the risk of injury associated with manned operations to inspect remote or dangerous areas, such as tunnels or underwater infrastructure, and enter previously inaccessible areas.”

Professor Stefan Williams from the Australian Centre for Robotics and School of Aerospace, Mechanical and Mechatronic Engineering said: “Australia faces a critical situation, with many of its post-World War II infrastructure assets rapidly approaching the end of their 50- to 80-year lifespan.”

“This ‘infrastructure cliff’ presents a mounting maintenance backlog, and the nation’s productivity and global competitiveness depend on efficient infrastructure networks. Our work has applications in public sector infrastructure, including roads and utility networks, but also in renewable energy, space, mining, and land care.”

The hub’s researchers said ARIAM will work with industry partners to develop robotic systems with unprecedented capabilities. Systems, including aerial, marine and legged robots, will demonstrate those skills in field trials.

“We will collaborate with Australian industry to tackle critical challenges within the robotics sector, ultimately paving the way for the development, manufacturing, and successful exportation of Australian robotics and autonomous technology to a growing global market,” Professor Williams said.

Image credit: iStockphoto/Ирина Мещерякова