Considerations When Starting With Augmented Reality
Once you have some ideas about how Augmented Reality (AR) can help your business (e.g., see AR Use Cases), you’ll quickly realize that you have numerous permutations of hardware and software decisions when getting started. Following are some points to consider as you choose which AR hardware and software to purchase.
Monocular vs Binocular
It’s cheaper to get glasses that only project to one eye (monocular), but you’ll find that creates an unnatural viewing experience that can lead to fatigue or headaches. It’s also impossible to provide a 3D experience without projecting to both eyes (binocular).
Of course, you can begin testing some AR application concepts using tablets or even smartphones. However, to fully explore and discover the unique potential or AR in the workplace, including but not limited to:
- Augmenting what you see
- Interpreting what you see
- Hands free operation
- Limitless viewing area (“screen size”)
- See-what-I-see collaboration
- Gesturing interface (breaking the bounds of keyboards, mice, and touchscreens)
Therefore, let’s focus on smartglasses (excuse the pun). If just starting your journey into AR, don’t spend too much. Buy a cheap pair to try out, right? Well, that leads to some real shortcomings that may cause you to underestimate the potential value of AR to your organization. For example:
Another reason not to skimp is for the user interaction. Inexpensive glasses usually will have something resembling a touchpad (e.g., a very tiny version on the side of the glasses) or mouse (e.g., a ring on your finger). These interfaces are usually cumbersome at best, and unusable for many uses (like drawing a circle around a broken valve).
A microphone is helpful for voice commands or remote collaboration as long as your use is in a quiet environment.
Some inexpensive glasses can recognize very crude hand gestures via their standard camera. Instead, higher end glasses will have a depth sensor that can enable hand gestures (remember the move “Minority Report”?).
Using gestures not only enables field workers with gloves or doctors with sterile hands, but using gestures similar to touchscreen gestures (swipe, point, etc.), is familiar for users, which should improve their acceptance of the device and perhaps even reduce their learning curve.
Similarly, you want glasses that can detect head motion, which can, for example, enable the user to view multiple “screens” floating in the air off to the side, by simply turning his/her head.
Smartglasses are computers needing an operating system. By far, Android is the choice for smartglass manufacturers, and there’s good reason for that. The compact size of wearables like smartglasses must rely on the component miniaturization driven by smartphones.
The open mobile standard with the most apps is Android. You’ll want to make sure you get a device that runs standard Android (not a proprietary forked version that may not run the apps you want). Keep in mind that you’ll want a device that can provide a usable touchscreen compatible interface or your users won’t be able to use the apps even if they run.
Microsoft Hololens runs a proprietary version of Windows that requires custom apps; that’s why they announce that only 80 apps work. The right Android smartglass should be able to run any of millions of Android apps. And the more existing apps and tools you leverage without “reinventing any wheels”, the faster you can recognize value from your AR devices.
In business, you may want to consider things like battery life. For example, will the battery last long enough for a typical work shift?
Or peripheral vision — In some environments, it may be safer to have a wider peripheral vision while wearing the device. Some devices block side vision. What ports does the device have? You may want to connect headsets, video output, etc.
What you’ll quickly realize after purchasing glasses is that it doesn’t do much without software. “But I’ll purchase an Android device with a touchscreen-compatible interface and run millions of existing apps, even my custom apps.” Sure, and that’s great if you just want to leverage the benefit of hands-free use. But you are unlikely to augment much reality from Play store apps.
“But I’ll purchase an Android device with a touchscreen-compatible interface and run millions of existing apps, even my custom apps.”
Sure, and that’s great if you just want to leverage the benefit of hands-free use. But you are unlikely to augment much reality from Play store apps.
Often with new computing devices, the first applications must be custom developed. And no exception, AR already has several SDKs like Metaio, Vuforia, and Wikitude. Of course, this still means you need to determine which SDKs will work on your device’s platform, meet the needs of your immediate and future use cases, and fit in with your organization’s application development standards.
Of course, this still means you need to determine which SDKs will work on your device’s platform, meet the needs of your immediate and future use cases, and fit in with your organization’s application development standards.
Fortunately, however, there are also a few “out-of-the-box” software platforms that can address some common enterprise AR use cases in:
- Field service and inspection
- Manufacturing and assembly
- Logistics and operations
- Medical and healthcare
For example, many of the use cases with demonstrable ROI include providing intelligent, contextual guidance to personnel out in the field dealing with complex machinery or situations. Others include capturing and documenting what the field person sees and does. Others still often involve escalation to an expert with “see-what-I-see” video collaboration. Thus, these software platforms can provide drag-and-drop creation of contextual taskflows that can guide a field worker.
Others include capturing and documenting what the field person sees and does. Others still often involve escalation to an expert with “see-what-I-see” video collaboration.
Thus, these software platforms can provide drag-and-drop creation of contextual taskflows that can guide a field worker.
Here are some considerations when comparing which out-of-the-box” software to choose
Finding software that not only meets your current use case needs, but also likely future needs is an obvious consideration
The good thing with software is that you can do just about anything. The bad part about software is that you can do just about anything…if you custom code it which is expensive (at least from TCO perspective).
Thus, the best of both worlds, if you can find it, is a product solution that can easily be extended.
Look for standard RESTful interfaces and the ability to develop add-in functionality. Not only does this allow you to customize as needed without being at the mercy of the software vendor, but you can potentially leverage other 3rd party add-ins if made available.
At some point, you may want to integrate with backend systems, whether pulling or pushing data (e.g., you may want to validate barcodes with valid material part numbers). Be sure to ask about capabilities for integrating data in either direction.
The easiest way to try a solution is if they offer a SaaS offering. In this cloud era, that should be easy to find.
However, if you want access to data or integration to systems behind the firewall, you may want to ensure the software vendor has an on-premises solution as well. Your business’ preference for OpEx or CapEx may also be a factor in your deployment choice.
Innovative technologies like AR move quickly. Therefore, look for software that supports multiple devices so you don’t get locked into one device. Also, for large organizations, it is possible you may have different devices used for different use cases throughout your company.
Many of the other points mentioned herein apply to enterprises. But also consider security and authentication. Can the software register and track devices and users, ideally tying into your corporate directory?
Total Cost of Ownership:
As you look ahead, you should keep an eye on costs, in fact, total cost of ownership.
Beyond licensing or subscription costs, this includes factors like professional services to get it installed, working, integrated, configured, maintained, and costs of internal skills also needed.
Unless you have unlimited time and money, you probably want to make choices that will deliver value or test your AR use cases quickly.
If you’re part of an innovation team discovering technology that might drive LOB innovation and have a healthy budget, you can probably purchase one of each category of device:
- one monocular and one binocular
- one with a crude user interface and one with high fidelity gestures
- one running standard Android and one running Windows (or a closed system)
But if you don’t have the budget to try “one of each” to fully appreciate the difference, trying just the cheapest device will probably not serve you well if you’re truly trying to discover the possibilities of AR. It might be money well spent to ensure you buy a device or two high-end smartglasses that can demonstrate the revolutionary benefits of AR.
To get value out of those glasses, look for a software (probably SaaS) off-the-shelf solution that requires minimal customization to address your target use case(s). Common examples include guided Taskflows and “see-what-I-see” collaboration with experts.
Often, enterprises entering AR will start with a proof of concept (POC) to test their use case hypotheses and measure initial ROI.
For the common AR use cases like field service, inspection, repair, manufacturing, logistics, assembly, and some medical healthcare, the ROI is virtually always clear (e.g., reduced downtime or travel costs, improved productivity or outcomes, etc.).
Thus, next steps lead to larger pilots before ultimate production deployments. Enough large enterprises have experienced sufficient returns to invest heavily in AR. Augmented Reality technology is evolving rapidly, so don’t get stuck in analysis paralysis. Above are some considerations. Above all, though, it’s best to get started.