




The display category can further be broken down into different areas of AR; each has its own benefits for different circumstances. Head-mounted display allows the user to interact and augmented the world around them while being fully immersed in it and still maintaining full mobility.
Handheld displays allow the user to still retain full mobility without having to immerse themselves into the AR; this is also the most likely candidate for mass production for consumers. Lastly spatial AR allows for use in the home, full interaction, limited mobility, and little to no immersion. For the most part this is using a webcam or some kind of camera and projector set-up. You'll find this type most commonly used now with online stores allowing you to print off a tag to "try" on their clothes to see how you like it.
For the most part all mobile computing power is referring to is the GPU or graphical processing unit. It seems kind of far strung and super high gadgetry but really for the most part it is a graphics card. Since the only computing needed is the coding and graphics being streamed to it, it is easier and smoother to use a GPU rather than a full computer system which breaks the work into two different chunks to send to the CPU and the graphics card, which then has to be put back together later down the road and cause a lag in the system between seeing a tag, recognizing it, and displaying the corresponding image.
Tracking is a matter entirely in and of itself. For the whole AR system to work it needs to be able to tell where it is in space and any movement deviation made from that point. This particular area is also where the greatest lag in AR systems is typically produced since most tracking systems being used at this point aren’t as sensitive as the systems need them to be. The best options right now would be one of the following: multiple GPS signals, differential GPS, real-time kinematic GPS, and the HiBall tracking system. As a personal suggestion, the HiBall tracking system seems to be the best bet at the moment for groups looking into experimenting with AR as its systems resolves linear motion of less than .2 millimeters, angular motion of less than .03 degrees, and an update rate of more than 1500 Hz with a latency of about 1 millisecond. Making it the best solution to the lagging problem commonly found in AR systems at this point in time.


I go to a whole University of like minded individuals in that aspect and we all have different areas of expertise [Game Design, Programming, Robotics, 3D Art and Animation, NetSec, Open Source] we decided to get together in a project group soon to turn business called ARinsanity. Got the punny haha part there?
Now at UAT to graduate we, among other requirements, we have to have a Senior Innovations Project. Something no one has done before that improves something or that is just really freaking cool. I myself prefer the SIPs that are both like the one I'm about to talk about.