14 Aug Expert Q&A: Capturing & visualising assets with Mark Merida
You’ve been with Geomatic Technologies for a long time. What have been some of the key changes you’ve seen in how asset capture is conducted over that time?
Historically asset capture was undertaken by large teams of field-staff with GPS, mapping software and tablets. This presented lots of challenges – safety, site access, logistics, data management, consistency and accuracy, not-to-mention cost and time. We are innovating and adopting new technologies at a staggering rate which is fundamentally changing the industry. 3D visualisation tools like AIMS3D and aerial mapping systems mounted on UAVs and other aircraft means our clients can access a digital-twin of their real-world network, quickly and cheaply.
Our rail customers, for example, can access asset condition from our high resolution imagery or clearance information to their critical assets from our LiDAR.
Your team have diverse skillsets – skilled technicians flying drones, photographers operating in remote locations and teams of programmers and data specialists. How do you coordinate them to create outcomes for clients?
We’re not just a software engineering company who are removed from the real challenges and realities of critical infrastructure. Our team’s expertise and experience covers the full gamut of what is required to understand client’s infrastructure challenges and deliver appropriate solutions. We pride ourselves on having this range of knowledge and capabilities in-house – from the engineer designing and manufacturing the camera rig on a 3d printer and the photographer sitting in a helicopter to the software engineer processing massive datasets and the designer testing the user interface, it all gets done here.
All our staff are encouraged to broaden their skill-sets. One week they may be capturing LiDAR data from a helicopter in rural Australia, the next they are processing imagery captured in North America. We find this approach helps us extend our offerings by providing a platform for cross collaboration of ideas.
You’re often capturing high resolution photos, video and LiDAR simultaneously – the amount of data must be very large. What are the challenges when dealing with this?
I read the other day that the computer that got Apollo 11 to the moon only had 64 kilobytes of memory which is astounding to think about these days. In any single day we are capturing terabytes of raw imagery and LiDAR data.
Traditionally, we processed data on our own infrastructure which had limitations – we now process all data in the cloud with Amazon Web Services. It was a challenging transition but the amount of data we can deal with now is really impressive. The speed we can move and process data means we can acquire data in the morning, upload in the afternoon and provide value for our clients the next day.
Your workshop provides a look at the bleeding edge of asset capture, photography, scanners, LiDAR, 3D printers and more but there are also some dusty corners that could almost serve as a museum of the various drones and other equipment – it seems like things move fairly fast.
It’s hard retire a piece of equipment that had a six figure price tag only a few years ago ago – but even though the the physical equipment may be taken out of service, the knowledge we have gained from configuring, testing and using it remains in the business. That experience feeds into the next project so that we are not only using the best equipment, we are using it at the peak of its capability.
It sounds like your team has quite a bit of fun working with new tech?
That’s definitely true – when we first got the 3D printers I was a bit sceptical about whether they would be helpful or just a toy for the crew to play with. After some some initial fun they quickly became critical to our work – we can now design, prototype, test and refine rigs, platforms and custom systems in hours rather than days.
What are the key technology innovations which will impact Asset Intelligence in the next few years?
The fusion of multiple sensors acquiring data from different platforms to solve complex problems is an area to watch. Driverless car technology is just around the corner. After that happens there isn’t much stopping us from mounting capture equipment on vehicles which are constantly scanning networks 24 hours a day, 7 days a week. Combine this with machine learning which automatically detects tiny changes in asset condition we really aren’t far off something that, until recently, sounded like science-fiction.
You could have a situation were a driverless vehicle captures an electricity network. That scan is immediately uploaded and analysed, which identifies where vegetation is encroaching and then a list of prioritised jobs are automatically sent to a mobile work crew – and that all happens within hours.
Rail clients are pushing for measurement sensors to be mounted on in service trains, reporting rail and rolling stock clearance defects back to a central control room to be prioritised and actioned immediately depending on severity.
The world never sleeps and all our services help keep critical assets operational with minimal maintenance times whilst maintaining the highest safety standards.