What’s Holding Back Augmented Reality in Construction
Augmented Reality (AR) is a hot topic—and there is a good reason for that. The technology has the potential to revolutionize all kinds of fields, from employee training to machine maintenance and even the way we build our cities. Yet despite all the neat headsets being released to the market, AR is not a perfect technology. There are still many things that need improvement. We sat down with Stéphane Côté, research director at building information modeling (BIM) software developer Bentley Systems’ Applied Research group, to talk about some of these issues keeping AR out of our construction sites.
Bentley and Augmented Reality
The research group at Bentley has been working on AR for quite some time—9 years, to be precise—and it certainly has a unique way of going about it. There was not much available in terms of AR headsets at that time, so the team used 2 webcams, an orientation sensor, a pair of goggles, and some duct tape to create its own AR system.
This MacGyvered system enabled the team to understand the advantages of AR in construction as well as some of the problems with it. With so many affordable headsets reaching the market nowadays, it is important to look at AR and consider what it can and cannot do for architecture, engineering, and construction (AEC).
Creating Reliable Information
AR has oodles of potential uses in the AEC industry. It can be used to demonstrate potential design options in the context of the real world or to supplant the need for construction workers to measure every little detail prior to installation. These “magic glasses,” as Côté called them, could be serious time-savers.
However, making mistakes in construction can be devastatingly expensive and if your information is not flawlessly accurate, you could be signing yourself up for a lot of trouble.
Why is that?
Côté used the example of a system that lets the user “see” through surfaces to locate infrastructure features like underground pipes.
“Let’s say the [AR] system displays the location of a pipe and the builders rely on that,” Côté hypothesized. “They start digging, they hit the pipe, and they have to evacuate the entire area. That is extremely expensive and they were misled because of the AR system.”
“It’s extremely important that the user of this [AR] system can rely on what he’s seeing,” Côté continued. “If it’s just a cool tool that displays something floating in the air but the user can’t rely on it for his work, he will just stop using it.”
So what is the solution?
“We have to [somehow] find a way to display the information to the user in such a way that he knows what he can rely on,” Côté explained. “For instance, it could be a color code. If the pipe is green, it could mean ‘Yes, that’s fine, you can rely on that. We surveyed that pipe last year, it is exactly at that location.’ If the pipe is red, it could mean ‘Okay, that pipe is that street, that’s all we know. It could be anywhere.’”
Until a solution is found, inaccurate information will certainly play a role in preventing mass adoption of AR on construction sites—and that brings up another issue.
Just Don’t Move
Once we get information accuracy up to snuff, there is something else we will need to consider: dynamic work environments.
No active construction site sits inert. They are places of constant movement, with workers buzzing around like bees trying to get the job finished as efficiently as possible. AR technology at this point is unfortunately best suited to static environments.
“Most of the time, augmented reality relies on the fact that the augmentation that you’re doing is on a static environment,” Côté explained. “For instance, say you have an office and you augment the office with a new wall or window. Of course it works well because the cameras that are installed on the headset will capture the scene and that scene doesn’t move—so the features that these cameras will detect are fixed.”
“If the room is changing, [for example] on a construction site and you don’t only have new walls being added but people carrying equipment and material, then the scene changes and the system has to rely on something that keeps changing,” said Côté. “I know researchers are working on that problem.”
There has been some progress on this front, especially in the context of testing AR in gaming environments, but it has not yet reached a level of reliability for high-stakes construction sites.
Check Your Blind Spots
Another aspect of AR systems that still requires more research is headset design. If you look at the HoloLens, for example, you see a large set of goggles that obscure part of your face and peripheral vision.
“We put a lot of importance on those nice headsets that enable you to view the world augmented, but as soon as you put something on your head, you block the view a little,” Côté explained. “As soon as you displace something in front of you, you don’t see what’s behind that augmentation.”
“Those things that you’re masking either on the side or behind the augmentation can be a safety issue,” Côté continued. “There may be a vehicle coming toward you or someone could be about to hit you with a piece of material.”
“We would like AR to improve safety on building sites and not decrease it, so somehow these augmentations will have to integrate well with the type of work being done and the type of equipment and type of clothes being worn so that it’s useful and not a threat,” said Côté.
There has been some work in this direction, including designs like AR contact lenses for bionic vision, but these designs will need to become fully realized before augmented reality is accepted on job sites. For more information about Bentley Systems’ efforts to bring augmented reality to AEC, check out Côté’s blog at http://tinyurl.com/hbzqget.
This article was published in the July 2016 issue of Insulation Outlook magazine. Copyright © 2016 National Insulation Association. All rights reserved. The contents of this website and Insulation Outlook magazine may not be reproduced in any means, in whole or in part, without the prior written permission of the publisher and NIA. Any unauthorized duplication is strictly prohibited and would violate NIA’s copyright and may violate other copyright agreements that NIA has with authors and partners. Contact firstname.lastname@example.org to reprint or reproduce this content.