Thursday, November 6, 2014

Virtual Reality Controls - The Key to Dominance is up in the Air

Thanks to the development of the Oculus Rift, Virtual Reality has gained another opportunity to become a mainstream medium. The Rift is receiving accolade after accolade and continues to be improved day by day. Based on my own experiences with the DK2, I wholly expect the Oculus Rift's eventual CV1 and other comparable HMDs to completely blow away the masses. Combine this with improved usage of headphones for immersive audio, and we get a combination that will provide a truly revolutionary sense of immersion. However, there is a big problem facing this VR wave that is holding it back. If this is resolved, VR won't just take a place in the market, it will take it over. That problem is the question of controls in VR.

To put it mildly, our current control methods are woefully inadequate in VR. The keyboard and mouse is rough at best due to the difficulty one can have even locating it with HMDs on and how easy it an be to lose your bearings. Controllers can be useful as well, but can be difficult to locate as well and isn't quite as versatile as many would like. To make matters more problematic, VR has a tendency to create a craving for more intuitive controls due to how well it immerses us. This is one area where none of our traditional gaming control methods are viable except in very particular usage scenarios. Creating an entirely new world only to limit the possibilities within it seems rather counter intuitive doesn't it?

Many people have recognized this shortcoming and are handling it in different ways. some people are designing under the limitations the traditional schemes provide, some are trying out the use of genre specific control items, and some are going about trying to create new devices or retrofit old ones to match the human body more accurately. I've found the systems being developed can be grouped together into particular groupings that can be useful in understanding their benefits and weaknesses. These groups are: Metaphor Based and Intent Based. Trigger Based systems use a device to obtain data based on the consequences of a user action. Intent based systems obtain data directly from the user's intended command, usually taking advantage of the source such as the brain or action potentials, or the unintended consequences of them.

Most of our current standard control systems are trigger based. Pressing the buttons on a controller, moving an analog stick, striking a pose for the Kinect, these all rely on a trigger that happens as a consequence of the user making one command to mean or translate to another. They are metaphorical in that one command is meant to stand in for another. The most advanced Trigger based systems we have, the motion controls of the Kinect and ControlVR, take advantage of the consequential movement of our bodies our intents make and attempt to map them one to one to our avatar to provide the illusion that our intents are directly being transmitted to the avatar.

Intent Based systems are considerably more advanced than Metaphor based systems as they cut out the middle man that is the metaphor and simply find a way to read the intent itself. There are VERY few systems capable of this, only two with consumer products to my knowledge really: EMG and EEG. EMG is rather tricky as it reads the electrical data that muscles create as they contract in response to motor signals. This is a rather fine line, being rather close to being a Metaphor based system since the muscle contractions must occur first. However, the muscles contractions themselves are not the intent users make so much as an out come that makes the user's perceived intent occur, which means the electrical data can be used to derive what the intent itself was and use that for the command. EEG on the other hand is much less nebulous as it reads the electrical data the creation of intents itself generate by observing the electrical data of the brain. 

If I had to say, the optimal system for VR is undoubtedly the Intent Based system as VR demands a degree of fidelity and a connection to the user only intent based systems provide. While this doesn't mean Metaphor systems can't work, I think they will have considerably more hurdles to pass over and blur the line between Augment Reality and Virtual Reality. They may be sister mediums, but there is a big difference in philosophy between the two that makes them incomparable in terms of what their most optimized to provide. In the mean time, I think using advanced metaphor systems to get the foundations in for VR controls, but some care should be taken to avoid creating systems that may become vestigial or antithetical to intent systems. Add in that intent systems have some overall incredibly powerful potential benefits, including negative latency where actions can be computed considerably before they would be expected to occur at all and the lack of the creation of unnecessary sensory noise in the best cases since an intent can be "overwritten" or negated before they can interrupt artificial data signals. 

Being that there are so many benefits, it's understandable that most people want this kind of a system as soon as possible. The primary problem however lay in that these methods of VR inputs are by far the most difficult to execute well. EEGs at this moment in time can be very slow due to the need to amplify, process, filter, and translate raw EEG data, which results tens of hundreds of milliseconds of latency. Add in to this that the method to get inputs is very rough due to the amount of processing and design necessary to create a usable, consistent data value. I think it's possible to do, but it will require a great amount of work to make consistently great.

Once the question of VR controls is solved however, I believe that the world will be faced with a medium so valuable, so versatile, so enrapturing, it demands attention and incorporation. Every aspect of our lives, from recreation to our workflows will be greatly impacted. Without a proper control system, VR will allow us to evolve our relationship and sentiments towards virtual content slightly, but it could still suffer the "gimmick" label since seeing VR content, while exciting, can be ineffectual towards certain tasks. I want this situation to be avoided at all costs, so the creation and subsequent standardization of a VR control  system is of the utmost importance. VR can succeed without it, but it can dominate with it. 

No comments:

Post a Comment