History of Remote Presence
John Markoff’s recent NY Times article presented current RPS offerings could be seen as marking the Big Bang of RPS. But the concept of telepresence (as I have defined it) has been around for some time—closer to 30 years, starting with Marvin Minsky’s contributions in the June 1980 edition of Omni magazine (recently reprinted in the IEEE Spectrum magazine).
From my research, I have put together a basic timeline of the mentions and creations that have been seen as remote presence / telepresence robots that I have captured. If I have missed any, please send me a note on the Contact page.
- 1994 – Personal Roving Presence (PRoP) at UC Berkeley
- 1999 – NurseBot Project from CMU/Pitt
- 2001 – BiReality at HP Labs
- 2002 – Companion robot from InTouch Health
- Oct 2004 – MILO introduced by RoboDynamics at RoboNexus International Robotics Conference
- 2007 – R.BOT 100 from R.Bot (Russian)
- 2007 – Ivan Anywhere at SQL Anywhere
- 2008 – TiLR introduced by RoboDynamics
- Jan 2009 – Anybots QA introduced by Anybots
- Oct 2009 – Texas One introduced by Willow Garage
- Nov 2009 – Anybots QB introduced at IREX by Anybots
Precursors to the remote presence explosion, these products attempted to focus on two-way visual capability and the representation of the pilot.
Robotists and other hardware hackers have been building variations on this theme for some time (see SuperDroid’s evolution of their version of remote presence system). At Willow Garage, the Texai came about when two robot engineers (Dallas Goecker and Curt Meyers) pulled together their research and various components from their machine shop.
Factors for Remote Presence
Then why now?
I believe there are a number of factors that have brought about the explosion of RPSes that have been the natural progression of technology and our ability to leverage it. Finally, the field of RPS has reached important tipping point. Below, I give a non-exhaustive list of what I believe are the main contributing factors (again, any corrections/suggestions – please comment below):
Ubiquity of Wi-Fi and large LAN pipes
Ethernet connections have been around on corporate LANs for some time, but Wi-Fi at tens-of-megabit speeds (802.11 a/b/g/n) have only recently become both ubiquitous and easy to use (see this IEEE article). And as 4G services are being deployed across the country, this bandwidth glut will increase pathways for video traffic.
Ubiquity of broadband connections
As corporations have upgraded their network infrastructure, staff has become used to the speed of response. Now, those same people are seeking the same network performance on their home connections. The demand for faster performance along with the increased entertainment and service offerings on the Internet have continued the growth of bandwidth to residences while bringing down the cost per megabit. (would love some data showing the graph of growth of broadband to the home and to the workplace)
Inexpensive video and computing power at everyone’s fingertips
While this may seem obvious, as recently as 2008, high-definition video processing on consumer devices was only found on high-end systems. Today, systems with separate graphics cards, lighting-fast CPU, large amounts of memory and incredibly quick bus and network speeds — all components that now surpass the need for normal usage and can handle heavier tasks like high-definition video compression—are available for less than $2K.
Solid improvements on video compression
Over the past 10 years, video technology has gone well beyond the original 15 frames-per-second (fps), Discrete Cosine Transform encoded Microsoft Media player to the high-definition video seen from Skype to Logitech. With the VP8/webM Project (an open video compression format by Google) video protocol standard and the HTML5 video tags providing the next potential jumping off point, how much longer until we see rich, high-definition motion video without jaggies and other artifacts?
You can see Willow Garage’s recent efforts on using VP8 on the Texai platform to provide rich, smooth video performance.
Increasing use of frameworks to build software platforms
Since the 90s, the use of basic coding frameworks that reduce development time through standardization and collaboration has been on the incline. Projects like the Player/Stage and the Robot Operating System (ROS) at Willow Garage are two examples of how frameworks allow for rapid development of prototypes while leveraging successes of others in previous work. The same phenomenon occurred with the progress on content management systems through the efforts with solutions like Drupal and WordPress.
There are also timely business incentives:
Cost and impact of travel
An example: Michael lives in Ohio but works and contributes to the team in Oregon. He knows the importance of being present in engineering meetings, since it greatly improves communication and his interpersonal relationships with his team members. Traveling from Ohio to Oregon is doable, but it’s also expensive and disruptive to his personal life. Transit time creates a productivity cost that far outweighs the benefits of his presence at the remote location.
With the above-mentioned factors and the impact of jetlag and familial strains, travel is not a positive contribution his performance – unless it provides a major breakthrough or a tremendous sale.
With every mile of air travel or car travel, the cost of fuel and the carbon contribution to the environment adds another tick in the box of costly use of our limited resources.
Increased work product fluidity
With increased power at people’s fingertips, work is becoming more fluid and digitized. In this day and age, it is rare to find a business, with external staff, than prefers traveling to a central location over telecommuting.
Increased competitive need for spontaneous collaboration
In our ever-competitive culture, collaboration and team building is part of creating successful products and services. There is also an increased use of work teams in common spaces (e.g. coworking) and an understanding that communications are more effective when people speak face-to-face. And since we are all human (or at least most of us are), the need for the face-to-face collaboration is far and away the most powerful input to the success of a team.
In this breakdown of factors, I have not included the improvements in manufacturing, logistics, telecoms (in terms of optical transport and back-haul), CNC milling machines, 3D printers or even the simple change of formerly hard-to-find hardware manuals now online.
I believe it’s only a matter of time before remote presence technology is developed into a feasible, marketable solution for many types of business; and as the IEEE Spectrum and NYTimes articles show, a number of people are well on their way.