AR/MR Smart Helmets for the AEC Industries: Flight or Fancy?

Smart Helmets: are they making us any smarter?

While BIM die-hards cling to their soon to be obsolete platforms, AR, and MR smart helmets featuring heads-up (HUD) and holographic displays are hitting the streets, and trying to make an impact on facilities management, and AEC technology in a big way – at least from an investment perspective, in DAQRI International’s case, to the tune of a whopping $200M they seek in R&D funding to develop their AR Smart Helmet™ (below, right).

smart helmets

If you think this is an unprecedented investment in an unproved industry, consider that Magic Leap, another AR head-gear developer raised USD$1.3B. What is the expectation of adoption for these devices in the AEC industries?

Roll back about one-year ago, when I demo’d the Google Project Tango, and posted my review. Then, we had a hand held device with on-board depth perception and area-mapping (self-learning) capabilities, as well as 3D depth sensing, accelerometer, thermometer, barometer, compass, GPS, and gyroscope that developers were taking to develop various applications that showed potential for the AEC industries.

“Relatively inexact sensors generate point-cloud images that are crude estimates of their tape-measured betters, at best.

The sensor arrays for the Tango and DAQRI Smart Helmet are surprisingly similar.

DAQRI’s Smart Helmet Data Sheet

from Google Tango Technical Specifications

Processor

NVIDIA, Tegra K1 w/ 192 CUDA cores

Sensors

Motion tracking camera, 3D depth sensing, Accelerometer, Ambient light, Barometer, Compass, GPS, and Gyroscope

Cameras

4 MP 2µm RGB-IR pixel sensor, 1 MP front facing, fixed focus

There are already dongles that sense and generate thermal images for the Tango platform, a feature that is built-in to the DAQRI HUD.

“Is the construction industry ready for space cadet architects bumbling around the job-site?

All of this was very impressive, though the AR rendering capabilities seemed somewhat crude, and far from obtaining accurate calibrations in any substantial way. In fact they still are. You can only squeeze so much photogrammetry into a hand-held device. For the nonce, this seems a moot point, as most of the development is targeting the gamer market, as one might expect.

Yet not so with MS Hololens, which shows greater promise as a tool in a much wider range of industries than any of the other smart helmets aspire to be. This, by virtue of its VR capabilities. SketchUp is partnering with MS Hololens to bring the MR arsenal to the AE industry with its SKetchUpViewer; however, SketchUp’s website blurb is less than inspiring.

hololens sketchup

HUDs & Smart Helmets Jumping the Shark from the safety of theory, to the chaos of practicum.

The first Tango Phone – the Phab 2 Pro (below right)was released to the public earlier last month, for about USD$499 by Lenovo. It’s still huge – over 6”, but considerably shrunk down from its former pocket-busting girth.

phab 2

and it is the first phone to use Google’s Tango platform. Google held a developer contest for the coolest apps, and what do you suppose they all were? –games, what else?

Also to be a feature at the CES show this year (it debuted last year) is the latest DAQRI Smart Helmet. The Smart Helmet has the same sensor arrays as does Tango, but the similarities end there. Smart Helmet uses its Heads Up display to bring sensor data into a mixed-reality overlay (of actual space) in the visor. This is the same HUD technology used in the automobile and motorcycle industry.

 

DAQRI will have competition at the show: helmets from Microsoft, Google, Intel, and even feckless Samsung are sure to attract major attention. Though they are both classified MR, Smart Helmet and Hololens do two different things:  Hololens is a mixed reality immersion overlay using VR holograms, whereas Smart Helmet overlays a 2D or 3D augmentation.

The Smart Helmet is intended to benefit facility managers, production managers, and plant managers by feeding real-time data to their HUD, from both field-sensors, as well as remote devices. Of course, you can gang the HUDs, which is an important asset. All contact point sensor data is fed to the visor on-board processor, which networks with the team. In addition, an AR image of a control panel can be overlain on the visor over the real control panel, and relay sequence of operations to the user in real time.

HUDs can bring up virtual touch interfaces and control panels that work with equipment sensors, and PLCs (project logic controllers)

Microsoft’s Hololens (below right) has much of the same AR capabilities as DAQRI’s Smart Helmet but its focus seems more design oriented, in a way that the Smart Helmet is not intended to be. Hololens is, by and large, AR, VR, and MR at the same time. It’s a VR immersive design collaboration tool that allows designers to see, and/or immerse themselves in virtual environments of BIM models in all the permutations that any design professional must negotiate every day.

“As with any new technology touted toward the AEC industries, ‘C’ is always the odd man out.

hololens

It’s unclear how much impact DAQRI and Microsoft think smart helmets will have for the building industry – as it remains nascent, at best. Rather, dynamic building sites are its weak-suit, and static design environments its showcase.

“…meanwhile, Back on Earth,

I have been the scheduler on a large hospital since 2014 where BIM models were part of the design package. On this project were some tens of thousands of preconstruction MEP clashes detected by the BIM system, mid-project. All had to be resolved by the contractors’ sketchers before the coordination could continue. If such a BIM model were uploaded to smart helmets it would resemble spaghetti, or a rat’s nest.

“Accordingly, the sketchers and detailers got together and worked things out on paper. Subsequently, digital issues followed; however, no BIM model as-builts: why would a builder need that?

The concept of using MR to navigate complex MEP coordination issues has some basic challenges that would have to be overcome before the technology could begin to be adopted on anything more than a negligible scale. In other words, smart helmets serve limited purposes of endeavors that have been traditionally done manually by detailers and sketchers creating shop drawings and overlays.

“Ground control to Major Tom: that hole in the slab is not virtual.

MR technology has the biggest appeal to the AE industry, and much less so to builders. This has always been the case with BIM, and for many of the same reasons. MR images rely on BIM models. Architects are expected to conduct pre-construction clash-detection between their 3D BIM models, and the engineers’ drawings, which are typically 2D dimensional, or single line. Therein lies a critical deficit in correlation that puts complex coordination issues out of reach.

For this reason, many preconstruction AE clash detection efforts are rendered ineffective – the biggest reason being the absence of volumetric MEP modeling (as opposed to linear) in their BIM overlays. As a consequence, the MEP coordination falls to sketchers and detailers of the respective trades.

“if your overlaying AR 2D images over 3D images onto a HUD you are going to be challenged to define conflicting program and available clearance and tolerances.

These drafters are seen onsite with their sketchpads, pencils, and Mylars during the MEP coordination phase, working out the kinks in single-line drawings. Once the MEP coordination is complete, the fabrication may begin. At this time, a BIM model of the coordinated work could be generated, and uploaded to a HUD, but then that would only be imaging for record, or as-built – in a word, superfluous.

There is no exigent need for a 3D BIM model in the construction sketching and detailing world, and by extension, no such need for MR smart helmets. Could building professionals avail themselves of the technology, there would remain the pervasive encumbrance of ubiquitous job-site and safety hazards that a space cadet would be dangerously oblivious to. The cost of procurement, implementation, and training will remain inaccessible until they become more mainstreamed.

Insofar as present adoption rates, the vendors have not been exactly forthcoming about sales volume or licensing. This is consistent with the typically secretive nature of R&D of the industry, but it does little to tell us how many design Stormtroopers are out there wearing them. Besides, the AE industry isn’t ready for the technology ipso facto they are still figuring out how to correctly use BIM.

Without mainstream adoption of BIM MEP models, HUD and MR MEP Coordination, smart helmets will become well mainstreamed in neither the AE, nor the C industries.

Written by

Derek Graham is a Primavera 6 scheduler and schedule oversight consultant for public and private sector projects with four decades of construction industry experience. He has been an active construction expert witness in over 30 cases - both nationwide, and Federally, since 2006, when his book Managing Residential Construction Projects, was published by McGraw Hill