Constants

roomXdim = 30     // x dimension of room
roomYdim = 30     // y dimension of room
roomZdim = 10     // height of room
lightBoxZ = 0.8     // displacement of the SearchLight from the ceiling
zLight = roomZdim - lightBoxZ     // height of SearchLight from the floor
xLight = 1     // displacement of SearchLight from room origin in x dimension
yLight = 1     // displacement of SearchLight from room origin in y dimension

Variables

xRef = (computer input)     // person's x location received via computer
yRef = (computer input)     // person's y location received via computer
xPerson = xRef - xLight     // person's x relation to SearchLight
yPerson = xRef - yLight     // person's y relation to SearchLight

Functions

thetaPitch = atan((yPerson) / (xPerson))     // degree of rotation across x/y
thetaTilt = atan(sqrt(sq(xPerson) + sq(yPerson)) / zLight)     // degree of        // rotation across (x,y)/z

Diagram

Rough diagram explaining code

Example

Let’s say someone was standing at the coordinates (20, 20) from the origin in the given room. What would be the angles at which the SearchLight would pitch and tilt?

xRef = 20
yRef = 20
xPerson = 19
yPerson = 19
thetaPitch = atan(19 / 19) = .785 (radians, or 45 degrees)
thetaTilt = atan(sqrt(sq(19) + sq(19)) / 9.2) = 1.241 (radians, or 71 degrees)

So according to the math as we’ve laid it out in the post, the SearchLight would arrive at a position 45 degrees from the origin across the x, y plane and 71 degrees from the origin of the z-axis towards x, y plane.

SearchLight Math (PDF)

Thomas and I are now working with Chris to prototype a tool for visualizing the presence of people existing in two separate spaces. As opposed to most popular tools out there used for communicating over long distances in real-time – tools such as web cameras, basic microphones, texting and instant messaging – the goal of this tool is to bring people out from behind the glass screens of computers, phones, and TVs and into our own rooms.

The way we’re doing this is by building a tool that can be mounted on a ceiling, which projects a beam of light onto the place equivalent to where a person exists in some other room either sitting or standing. To demonstrate the concept, we’ll place a Kinect in room #1 to track the location of a person in that room, and we’ll place a pan-tilt flashlight mount onto the ceiling  in room #2 to project the location that person.

The pan-tilt flashlight mount has been built over the past couple days with PVC pipe, foam core, servo motors, and an arduino (see below). Thomas just got the arduino to work with the servo motors, so we’re moving onto the next part of the project – the Kinect Sensor.

The "SearchLight" & Kinect

Using the Kinect Sensor we’re trying to find a reliable way to track a person’s position in a room. We’re using OpenKinect - an open source library of code for processing and manipulating data from the Kinect – to experiment on the possibilities. As great as the OpenKinect project is, no one has developed the skeleton tracker that Microsoft includes with its own SDK for the Kinect, but we’re exploring other ways to reliably track postion, such as through a facial recognition algorithm. Stay tuned!

Thomas doing math

Here is a video that I always think of as being a really cool example of how powerful spacial sound can be. Put on some headphones, press play and close your eyes. Extremely fun for a youtube video:

This NYTimes article is a good example of how strong a desire there is for a good solution in this area: New York Times: Skype Postpones Sweet Sorrow

I’ve spent the last few weeks mapping out existing telepresence offerings and their history, conducting a range of interviews and reading a variety of research on the related field of virtual reality.

I’ve been reading Infinite Reality.  It’s a bit light but what has been interesting are the virtual reality fMRI studies that have been conducted. I’m fascinated by how quickly and readily users will have physical reactions to events they rationally know to be illusions. This could be important in a variety of ways, the most obvious being that lower fidelity avatars can still be very effective as long as they are used correctly.

I have found myself fascinated with the research on how behaviour and perception changes based on different modes of communication and incarnations of avatars.  I have been inspired by some research at MIT Media labs on the MeBot project. I don’t think I want to go this route (avatar robots) but it shows how powerful working prototypes can be for exploring a new form of communication.

I have also conducted interviews with employees at international design firms and couples in long distance relationships. Both of which have been extremely interesting. These are two very different problems to solve for each of these groups.

Many think these problems will naturally be solved when bandwidth is sufficient. I suspect that there is a great deal of design that could create a far better experience in the near and distant future. I generally feel that further bandwidth gains alone will only solve a minority share of the existing problems. In my thesis proposal presentation I made the analogy to video games:

We can either continue to throw bandwidth at this problem

or we can create an alternative user experience. I think one of these could be very disruptive.

• Poor experience is tolerated
• Little emotion is conveyed
• Communication is anything but natural

This reminds me of how people thought of blackberry’s in 2006. It was pretty neat that you had your email on the go, but most people didn’t even realize just how horrible their experience was.

Telepresence is in need of the same iPhone launch like kick start in order to make it clear just how hackish current telepresence is. The innovation that needs to happen will not be in slightly higher quality video feeds, but in a completely new way of thinking about communication and what it means to be with someone who is far away.

There are existing solutions. Some are for business, and I haven’t encountered any that seem enjoyable at all. They technically get the job done, but all fun seems to be lost in the process as so often happens in business focused products. Almost every other product is a surprisingly flawed consumer solutions based purely on the fact that there happens to be a webcam on top of your laptop display.

I like some more than others. Google Video seems robust but poorly designed, Skype has good voice quality but drops the call constantly. FaceTime doesn’t even need the application to be open inorder to receive a call and frames the video so the subject is really the only thing in the shot.  These are good iterative ideas, but small changes will not open up an extremely limited form of telepresence.

I often talk about these solutions as the prison visitation call. You hunch over a computer and speak with the other person.  You can see their facial expression, but the information conveyed is pretty limited beyond that.

Over the coming weeks I need to figure out what drives people to use video calling. I’ll try to find out what they’re missing in their long distance communication and see how many of those problems can be addressed by rethinking current technology.

I think this opportunity has been validated by what we’ve seen recently with Microsoft purchasing Skype. There is a large group of people who want the full immersion experience that video chat is starting to offer.   They can accomplish collaborative tasks faster and with a greater connection than they could via text only solutions. I personally think we are on the edge of a wave of what people think of as communication on the web, both in 1 to 1 and many to many. I believe a significantly larger share of the face to face connection that salesmen and grandparents alike point to when justifying their travel costs can be achieved digitally. Though certainly nice, I don’t think that higher resolution video is the whole solution. I intend to investigate the true nature of problem solving/distance communication and begin to address the unmet needs I find to create a much more engaging experience.

This final project will consume the next 3 months.  I hope to use this blog to share and chronicle my thoughts and any developments.