Mobile Telepresence Design

I was introduced this week to a new concept, tele-presence robotics, via the NYTimes and PC World.
First, the article, The Boss Is Robotic, and Rolling Up Behind You, interesting, slightly weird, but likely the dawn of something much larger.

Mr. Beltzner rolls the robot to a large conference table in the Mountain View headquarters of the Mozilla Corporation, maker of Firefox, a popular Web browser. By swiveling his camera eye back and forth, he can see the entire room and chats comfortably with the assembled team.
An hour earlier, Mr. Beltzner, director of Firefox, was logged into a different robot on the other side of the building to attend the weekly all-hands meeting. With a pink lei on one shoulder and a jaunty cap on the other, the robot was surrounded by more than 100 young software engineers, each sitting with a wirelessly connected laptop.

Aside from the occasional greeting, no one seems to notice the disembodied Mr. Beltzner until he is called upon by Mary Colvig, a Mozilla marketing manager. She wants employees to share the chore of leading tours of the office each week. …

All the science fiction I’ve read certainly prepared me for this. Excited, I thought there must be some cool looking hardware, handling the delicate territory of personal interaction with tech. Nope, disaster. Seldom is the gap between the design I imagine from the description of the usage, and the shock and disappointment of what has actually been built.

Youtube video from PCWorld
Looking into it a bit more, I find the more scholarly IEEE Special section on telepresence

Even worse the descriptions of usage, falling over, batteries dying, difficulty steering. Not only is it ugly, it can seem laughable in practice.

This is the area in which I thrive however, that gap between what something could be and what it is. So out comes the sketchbook. Try and sketch some solutions. The uncanny valley theory quickly comes to mind. Designing the elements needed into something that doesn’t look like a Dalek or a vacuum cleaner is tough. I imagine different systems of locomotion, suspension from the ceiling, mobile laser projections on surfaces, rejecting all of the imagined appearances. Time to break out of the the box. The most complex part of the system is the robotics, the communication system is glorified Skype. Moving a wireless unit around a room should not be that difficult. Eureka, sneaker net. Embrace the oddity with both arms.

Why not use the most mobile, adaptive, and readily available system possible? The human transporter. Since the object is to transport the likeness of a human face for communication, the arrangement becomes obvious. A head mounted unit. The world is populated by possible avatar face couriers, senior citizens, unemployed, performing arts majors, etc.

Teleface is born. A head mounted display unit with camera and 2 way sound transmission, wirelessly linked. Everything is integrated into a helmet-like unit that can be worn by virtually any operator. The entry level costs drop by more than 60% as complicated robotics systems are replaced.

The teleface operator sees interface and environment through stereo glasses on the reverse side of monitor. Headphones transmit audio instructions from the remote operator via web interface, as in current systems. The teleface operator hears the instruction in his or her own language, right left forward, sit, lean forward, cross arms, etc. During conversations between the remote operator and communication intendee, the headphones play music to mask the conversation from the teleface courier. Privacy is assured. Additional features in the internal stereo display, will allow the teleface operator to mimic the body positions of the remote operator. Empathy and NLP come into play, performing arts students bring value to the table.

Now this is a guaranteed shock, it is necessary through careful design to minimize the shock. White colors and soft lines minimize the threat. The black eye covers communicate “sunglasses” with an undertone of blindness. Again the communication of privacy and discretion. A white jumpsuit is recommended as attire for the teleface operator as it communicates a blank slate as well as a workman aesthetic. For medical application generic blue or green scrubs will provide the same function of disguise.

Feedback and discussion welcome.

Maybe it is more of a medallion, the way Twiki carried around that other AI brain around his neck in Buck Rogers:

Twiki! It’s been a while. (Always reminds me of Dr Smile in “The Three Stigmata of Palmer Eldrich” by Philip K Dick.)

The medallion is a valid concept, certainly design and hardware wise a good location. Two psycho-socio-logical issues with that, the first is the “tele courier’s” face sitting across the desk from you. Hard not to look at a stranger’s face, situated directly above the person you are having the tele-conversation with. The second issue is the height of the display, for social interaction it should be at head level.

The concept attempted here, is intended more akin to Japanese Bunraku puppeteers, dressed in black, plainly visible to the audience, but behind the main object, ignored and anonymous.

Do you think it is one of those issues that doesn’t demand a physical thing at all. Maybe the boss will just open up a video chat out of your control on your desktop, laptop, or company paid phone via a piece of software. Maybe there doesn’t need to be a physical thing, the way that there was never really a “video telephone”, just a piece of software called skype or v-chat and now face time (I never use face time BTW, but I frequently use v-chat and skype)

I thought so initially also, after reading more of the first hand experiences with telepresence I see a deeper trend. The physical aspect adds another richer layer to the interaction and to productivity. In my self projection into the imagined environment there are more possibilities by being in motion around an office or factory than just peering at someone sitting at a desk through their monitor perched webcam. The surveillance aspect is an important one, perhaps the physical moving presence has an acceptable quality, whereas the sudden, immediate video eye overseer would be too stressful. I am not familiar with the psychology of working under constant or random surveillance, the teleface system makes the remote presence one to one with physical presence. (The interface control could have a “sneak” request for the operator to walk quietly up to the intended target, not a great morale booster though.)
Article with insights by RoboDynamics founder Fred Nikgohar

Since they first started to field TiLRs back in 2008, RoboDynamics has been collecting data on how people use and react to telerobots. These case studies have been remarkably insightful. Here are some interesting results:

  • In an office where a telerobot is operated by a single user it takes about 4 days before coworkers start to refer to the robot by the operator’s name. It’s not, “where’s the robot?” it’s “where’s Jeff?”
  • About 30% of operating time is spent driving the robot around. Nearly 100% of users, however, didn’t want this time to be reduced. Driving around time was great for walking conversations, thinking, etc.
  • About 68% of conversations are started on the robot side. Often a coworker walks away from their own desk, seeks out the robot, and talks to the operator. This is in spite of the fact that the operator is easily accessible from the coworker’s desk via email, phone, or even video conferencing.
  • After a week, 61% of operators say they would be severely hampered if they had to go back to telecommuting in without TiLR.

    Want an example? Let’s examine the case of the $3000 work day. Nikgohar had a customer that had to travel a lot. When he was in his factory workers produced about $3000 worth of product per day. When he was out they produced $2000. Makes sense. When the boss is away you take longer breaks, you spend more time socializing. Anyone who has worked long enough has probably seen this effect. So now the customer gets a TiLR. When he is away, he starts to commute in via the robot and drive around the factory. His hope is that his virtual robotic presence will recoup some productivity. Even if its only up to $2500 or so it will be well worth it. What happens? When he’s out of the office and using the robot productivity does go back up. But not to $2500, or even $3000. Suddenly the factory is producing $4000 a day. What the hell happened?

I recall a graph (edit: Allen Curve) showing the incidence of communication in an office falling off rapidly and down to zero if the individuals were located more than a certain distance apart. The physical proximity of a co-worker and the green icon of availability on my Skype dialog are not equal, even after many years of using it.

Tom Allen, the dean of researchers on the work behaviors of scientists and engineers, found more than two decades ago that technical workers (a proxy for knowledge workers) whose desks are more than thirty meters apart have a frequency of communications that is roughly zero. Some might argue that e-mail and instant messaging have changed the relationship between physical proximity and communication. However, I’d argue that you rarely e-mail or IM intensely with someone you don’t know. Assuming it’s still true, Allen’s important and oft-cited finding means that companies should design work environments so that knowledge workers who need to communicate are physically close to each other. Of course, this requires some strategizing about who needs to be talking with whom. Organizations such as 3M and Herman Miller have tried to do just that in the design of some of their facilities. > Why Office Design Matters - HBS Working Knowledge - Harvard Business School

I don’t like defaulting to the “there’s and app for that” but I do notice how much the younger creatives in our office text to one another all day to communicate. There might be a generational variable here. Also, I’m on a work computer and I assume every keystroke is logged somewhere and looking for inappropriate combinations. In a way, that is the beginning, most of us are either oblivious to this or accept it, and I wonder how that codification of digital privacy will affect us in 10 years, 20…

Again, I go to the video phone. Most offices have at least one $30k+ video conferencing set up that no one uses, but they will use v-chat. I think those devices can be predictive and then fall of a slope of expectation into a trough of disillusionment.

Right now the issue you are getting at is essentially handled by middle management. ie CEO sets goals and targets, VP’s figure out strategies, Managers develop day to day specific orders and monitor progress.

I agree with the need in this domain, but your concept and headgear are likely to freak people out. They remind me of two dystopian science fictions:

Conceptually it reminds me of the “meat puppets” in William Gibson’s Neuromancer–people who submit their bodies to the control of others. AVATAR is also based on this concept.

Aesthetically it reminds me of the helmeted “enforcer robots” in Lucas’ THX 1138:

Good points, my challenge here is to make a few very harsh points and the conceptual freakout, softer. This is dystopian domain for certain, I am trying to make it palatable in a social way, and find a design solution. The usage in the first renders of a solid white background make it emotionally cold and lend comparisons to sterile futures, that is an error.

The person that delivers the packages in a large office or runs the coffee cart is no less of a robot, he or she simply has a face. There are issues with making the “operator” feel like they are doing something of value, and the users of the system to feel comfortable with the inert third party in the middle of the process of communication.

Regarding materials, perhaps white can be eliminated as well as black for being reminiscent of police helmets. Here is a more contextual example with some material changes.

Maybe it is more of a medallion, the way Twiki carried around that other AI brain around his neck in Buck Rogers:

Or simply …

Lew, I was looking forward to your insight. I expected at least an image from Sleeper! I am taking the position almost of a lawyer here, arguing a case to its logical or perhaps illogical conclusion.

The best method in 2010-11 to meet the needs of the telepresence function. Right now this is the main contender:
Anybots, your personal avatar

Designwise I find it to be lacking in critical areas, appearance, flexibility, smoothness of movement, facial communication. Watch their main page video and I’m sure most industrial and experience designers will be left wanting more. As a concept however, I find it intriguing. As for the “teleface” design solution of hiring a person to take the place of a robot, a controversial avenue I want to explore.

A very interesting challenge you’ve got here.

You brought up The Uncanny Valley Theory early and I think your exploration is touching upon a similar type of reaction from a different angle. In the Uncanny Valley as things approach human resemblance without fully realizing that resemblance they seem strange and alien. Here you are essentially taking an actual human and replacing their normal “interface” (which probably the most important physical elements of that person’s identity) with a screen and a remote person. This effectively removes some of the person’s inherent humanity, but further by replacing these elements with an electronic interface, I think another type of uncanny valley begins to form. There is also this surrogate that I have some primary / physical engagement with, but is not meant to be my primary interface? It’s an intriguing problem for sure, but I’d be interested to see a lot more breadth of exploration. I do appreciate your thoughts and taking on this admittedly controversial point of view.

Keep us posted!

Okay, time to abandon the Uncanny Valley or the Valley of the Stepford Dolls. I realized the lingering ill effects of HR Giger in the use of the face as the communication carrier. Technologically possible, solveable and intriguing as an experiment, socially impossible, of course. So, remove the controversial aspects of human transport and concentrate the natural aspects.

The mobile telepresence concept still holds a great deal of interest for me. As the current state of robotics is nowhere near able to negotiate the real world environments that telepresence can explore, so I am staying with the concept of a human as the transport system. Flexibility,ability to fit into social and structural systems, security, etc.

Here is generation two of a mobile telepresence concept. Thin, two sided placard shaped system that places the display at the height of a human face and allows a transportation from location to location or a physical inspection of a site.

Front view.

Rear view showing sketch of user interface with SteadiCam style pivot and handle.

The lower base allows placement on a desk or meeting table, can be deposited and left without operator continued attention. The location of the handle gives the same natural human stabilization as a SteadiCam mount to make life less nauseous for the remote tele-operator. Smooth video. The system uses a set of dual screens, one for the communication system and one screen on the reverse for operator instruction. Due to the requirements of instruction, an e-ink system would fit the requirements and have lower weight and energy usage. Chargeable through setting on a base. Camera allows for independent rotation by the remote operator.

Construction site tours, remote visitors to patients in hospitals or instututions (rentals), meetings, virtual factory tours for international business prospecting and quality control. More to come.

Adapted the base to similar language as the TeleFace placard, added the stereo speaker holes.

Defensive publication checklist.

  1. Dual screened, front and back telepresence mobile placard.
    Mobile wireless communication device for easy transportation, used as a remotely directed system for simulating presence in a distant, difficult environment.

Batteries and electronics are housing in the “foot” section to provide a counter balance to the top screen section and stabilize.

Pivoting handle allows the unit to self stabilize, as in Steadicam camera technology, to help eliminate sudden changes in the angles of the transmitted video which are disorienting to the remote viewer.

Swivel camera allows the remote user to alter the viewing angle independent of the main unit. An alternative would be to capture a full fisheye field of view and use software to focus on a visible section and “unwrap” the fisheye distortion.

Augmented reality possible in both displays. The rear side, used for guidance by the operator, can be overlaid with the known environment. The transmitted view can have local information, for example inside a factory, construction site, hospital, cholera epidemic site, etc, augmented into a layer over the live video stream.

The unit houses transmitter and receiver, GPS and directional sensors and electronics and systems required to run.

A web type interface would be used by the remote operator to direct the path of the TeleFace unit.

A friend just shared this and I was reminded of this thread.

It may turn out to be as simple as “appification”. Simple and discrete devices that augment the world around us seamlessly…

For some consideration:

If you have an iPhone/Pad, download Mojo Masks and give it a whirl, you’ll see the possibilities. Awesome stuff.

Cool, but even better, projection. Have given this a lot of thought around Halloween when the first microprojectors came out. Have not actualized it yet, fortunately someone else has.


- YouTube This, unfortunately was what the seventeen year old me thought was cool.